Comprehensive final

Lakukan tugas rumah & ujian kamu dengan baik sekarang menggunakan Quizwiz!

Which of the following correlations indicates the strongest relationship between two variables?

+ .65 -1.45 + .03 -.91

The typical correlation coefficient varies from

-1.0 to +1.0

Which of the following correlations indicates the strongest relationship between two variables?

1.43 0.54 0.04 -0.94

Chapter 6 Correlation Methods and Statistics Topics

1.Conducting Correlational Research 2.Magnitude, Scatterplots, and Types of Relationships 3.Misinterpreting Correlations 4.Prediction and Correlation 5.Advanced Correlational Techniques: Regression Analysis

TEN COMMANDMENTS OF INTERVIEWING (Survey Research)

1.Do not begin the interview cold. 2.Remember that you are there to get information. 3.Be direct. 4.Dress appropriately. 5.Find a quiet place to conduct the interview. 6.If your interviewee doesn't give a satisfactory answer the first time, don't give up. 7.Use a tape recorder. 8.Make the interviewee a part of the interview. 9.Practice. 10.Thank interviewees for their help, and ask for questions. Source: Salkind, Neil J. (2003) Exploring Research, 5th Edition. Upper Saddle River, NJ.

In a 2 x 2 x 2 factorial design, there are how many levels of each independent variable?

In a 3 x 3 factorial design with repeated measures on the last factor, how many independent variables are there?

In a 2 x 2 x 2 Factorial Design, there are how many independent variables?

In a 4 x 2 x 4 Factorial Design, there are how many independent variables?

In a 3 x 3 complete factorial design, how many different experimental conditions would there be?

People who are of different ages can also be of different generations and thus differ in their attitudes and experiences. This can be a disadvantage in research that tests people of different ages. What is the term for this effect?

A cohort effect

Which of the following would NOT be found in the Method section of an APA manuscript?

A description of the statistical test used in the study

Which of the following would be found in the Results section of an APA manuscript?

A description of the statistical test used in the study

Response bias refers to

A habitual way of answering questions

Which of the follow proposals would require a full board review by the University of Oklahoma IRB?

A laboratory study utilizing older adults (age 70 or above) and testing their memory

Steinka-Fry, Tanner-Smith, and Grant (2015) wanted to investigate whether an intervention targeting college students turning 21 reduced the amount of 21st birthday drinking. They found nine research studies on this topic and they combined the results from these nine studies to determine that overall, although those who received the intervention had a slightly lower blood alcohol content than those who did not, there was no difference in the amount of alcohol consumed. What kind of study did Steinka-Fry et al. do?

A meta-analysis

Which of the following research methods would not require informed consent for participants under most circumstances?

A naturalistic observation study taking place in a public setting

Finding Relevant Literature: An Important Distinction

A primary source is a complete research report about a particular study or studies the author or authors conducted. It includes sections on participants, method, and results. A secondary source does not provide the complete research report of a study or studies; it provides only summaries of the cited works. You can read secondary sources as a good first step in learning about your topic. They can give you an overview of an area and ideas about what primary sources to investigate. It is important to read primary sources when you are preparing a literature review so you can know the details regarding how authors completed their research.

there is no linear relationship between the variables r = 0

A researcher runs a correlational study and finds a Pearson's product moment correlation of r = 0. She can conclude that

she must have calculated her correlation incorrectly r = 1.12

A researcher runs a correlational study and finds a Pearson's product moment correlation of r = 1.12. She can conclude

In a factorial design, there are 2 independent variables: Ink color and word difficulty. The results show that difficult words written in red ink (mean = 8) are remembered better than difficult words written in black ink (mean =3), but ink color didn't make a difference for easy words (mean of 8 for both conditions). Based on this finding, we could expect that the ANOVA would show

A word difficulty x ink color interaction

In a factorial design, there are 2 independent variables: Ink color and word difficulty. The results show that difficult words written in red ink are remembered better than difficult words written in black ink, but ink color didn't make a difference for each words. Based on this finding, we could expect that the ANOVA would show

A word difficulty x ink color interaction

Yes

According to the APA ethical guidelines, it is acceptable to use deception in a study

Experimental Design with One Independent Variable: Analysis for One Independent Variable with More Than Two Groups: A Sample Write-Up

An ANOVA revealed that the music did impact the time it took the participants to place a bet, F (2, 257) = 30.07, p < .001, r = .71. •Again, details at the end of the sentence provide information regarding the statistical test that was performed (an "F" test refers to an ANOVA), and the numbers in parentheses refer to the degrees of freedom for this test. The probability value indicates the likelihood that this result is due to chance, and the "r" here is the effect size. •This sample sentence just indicates that the independent variable had an impact, but we need to use a post-hoc test to determine which groups were different from each other (see next slide). The post-hoc test revealed that those who listened to fast tempo music had faster betting speeds (M = 1.49, SD = .04) than those who did not listen to music (M = 1.98, SD = .46) and those who listened to a slower tempo of music (M = 2.53, SD = .39). In addition, the betting speed observed during the slower tempo music was significantly slower than the betting speed observed during the condition with no music (the mean of 2.53 was significantly different than the mean of 1.98). •The above sentences represent comparisons of the different groups with the means and standard deviations noted.

randomly assign 50 to the therapy group, and 50 to the control group.

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. In the study of the new therapy, what would be the best way to decide who is in the experimental group and who is in the control group?

whether the participant received the therapy or the control independent variable

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In this study, what is the independent variable?

a type II error I fail to reject my null hypothesis and conclude that the new therapy has no effect on depression, when in reality it does lessen depression

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: I fail to reject my null hypothesis and conclude that the new therapy has no effect on depression, when in reality it does lessen depression. I have now made which type of error?

a type I error I reject my null hypothesis and conclude that the new therapy reduced depression symptoms, when in reality it has no effect

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In this study of the new therapy, I reject my null hypothesis and conclude that the new therapy reduced depression symptoms, when in reality it has no effect on depression. I have made which type of error?

the number of depressive symptoms in each participant at the end of the study dependent variable

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: what is the dependent variable?

solve practical problems

Applied research is done to

Which research technique would usually have the least amount of reactivity in particpants?

Archival research

Stratified sampling

Assume that a survey on attitudes toward hopeful U.S. presidential candidates is being conducted on our campus. The distribution of male and female participants from the campus is carefully matched to the distribution of males and females in the general population of U.S. citizens. This is an example of ___.

Which of the following types of research would most likely rely on a small-n design?

Behavior modification in an institutional setting

SIgmund Freud built his theory of Psychoanalysis by using a detrailed record of a few patients of his. This is an example of which type of descriptive method?

Case study method

Major headings in an APA article, such as "Results" and "Discussion" are usually ___ in the copy manuscript.

Centered headings

Nominal

Classifying students as an English major, a Business major, or an Education major would be using which level of measurement?

Descriptive research focuses on

Collecting observations on behavior

Which sampling technique is would be considered nonprobability sampling?

Convenience

The Goals of Science

Description-describe the phenomenon of interest. Explanation-explain the phenomenon of interest. Prediction-predict when the observed phenomenon will occur again. Control-control the phenomenon of interest.

Collecting observations on behavior

Descriptive research focuses on

strong positive correlation

Dr. Shapiro has developed a test for shyness. He administers the test to a sample of respondents twice, a week apart. He then calculates a correlation coefficient to determine how well the scores at the two testing session match. What kind of correlation coefficient does he want?

Asking an expert to look over your new scale of Extraversion to see if it appears valid would be used to establish ________ ?

Face validity

A between-subjects (or between-groups) design uses the same participants in each treatment group?

False

A between-subjects (or between-groups) design uses the same participants in each treatment group? False

False

A within-subjects design uses different participants in each group.

False

If a study has a significant interaction, this will make the researcher emphasize the main effects even more than if the interaction wasn't significant.

False

If my correlation coefficient is statistically significant, it means that I definitely have a strong relationship between the two variables.

False

If my correlation coefficient is statistically significant, it means that I definitely have a strong relationship between the two variables. False

False

If my experiment has a confounding variable, it will increase the interval validity of my study. False

False

In Latin Squares counterbalancing, every possible order of treatment is used in the experiment.

False

Informed consent is always required in a psychological study

False

It is possible to have a factorial design with only 1 independent variable.

False

It is possible to prove both the null hypothesis and the alternative hypothesis in an experiment.

False

It is possible to prove both the null hypothesis and the alternative hypothesis in an experiment. False

False

It is possible to reduce both the chance of making a type I error and the chance of making a type II error at the same time in a single study.

False

Using counterbalancing will prevent participants from having practice effects or fatigue effects in the experiment.

False

Random sampling means that each member of a particular population

Has an equal chance of being selected

When I say that my study has a significance level (or alpha level) of .05, it means that

I have a 5% chance of making a type I error

Prediction

If the data were not consistent with the hypotheses, and the hypotheses were derived from a theory, then the theory needs to be modified. Once we modify the theory, we could generate new hypotheses and test them. Each time our hypotheses are supported, we gain confidence in the theory.

Prediction

If the results are as we hypothesized, we say, "The data support the hypothesis." (singular) or "The data support the hypotheses." (plural) If the results are different than what we hypothesized we say, "The data did not support the hypothesis" or "The data did not support the hypotheses."

Ordinal

If you classify clothing sizes as small, medium, large and extra large, you are using which level of measurement

Varied by the researcher independent variables

In experiments, independent variables are

Ratio

In my study of a new therapy type, I measure the number of symptoms of depression (out of 9 in the DSM-IV) that each participant has at the end of the study. My dependent variable is using which level of measurement?

False

Informed consent is always required in a psychological study

The Fahrenheit scale of temperature uses which scale of measurement?

Interval

When you are writing a manuscript, where do you develop your rationale for doing your particular research project?

Introduction

Which of these sections comes first in an APA style manuscript?

Introduction

Is it possible for a psychological measure to have low reliability but high validity?

Probably the most common limitation associated with correlational research is that

It does not prove causation

Probably the most common problem associated with correlational research is that

It does not prove causation

Which of the following situations best illustrates the principle of intuition?

Just knowing in your gut that "birds of a feather flock together."

If you compared the problem solving ability of a group of six-year olds to their own performance when they were eight years old, your design would be

Longitudinal

Description

Many psychological researchers describe observable behaviors (behaviors that can be seen). We observe these behaviors in a systematic manner and document the results of our observations. If we are interested in describing non-observable phenomenon (such as thoughts), we can design questions to ask participants what they think and how they feel

Control

Many psychological researchers have the goal of knowing how to influence or even control attitudes and behavior (for example, learning how to control aggressive behaviors in children).

validity

My new measure of intelligence actually measures reading comprehension instead of intelligence. We would say that this measure is low in

Which of the following research methods usually has the highest ecological validity?

Naturalistic Observation

Which of the following research methods usually takes place in a natural setting?

Naturalistic Observation

Is it possible for a psychological measure to have low reliability but high validity?

Finding Relevant Literature

No matter where you get your research idea, before you plan your study, you should investigate what others have written that is relevant to your topic. To locate sources relevant to your topic, you'll likely use PsycINFO, a computerized database that provides citations of books and journal articles, journal article abstracts and in some cases, the full text of journal articles. Ask PsycINFO to limit your search to a specific "publication type." Ask PsycINFO to limit your search to a particular "age range." Ask PsycINFO to limit your search to a particular "population." Use the Boolean operator "and" to narrow a search (for example, "alcohol abuse" and "aggressive behavior"). Use the Boolean operator "or" to broaden your search (for example, "sexting" or "cybersex"). Use truncation. For example, the truncated phrase "aggress*" yields sources that have the words "aggression," and "aggressive" in a search field.

Classifying students as an English major, a Business major, or an Education major would be using which level of measurement?

Nominal

Which of the following situations best illustrates the principle of empiricism?

Observing the friends you have to see if they tend to form relationships with people who are similar to them.

confounding variables

Of the two sources of error, which is more serious-extraneous variables or confounding variables?

If you classify clothing sizes as small, medium, large and extra large, you are using which level of measurement

Ordinal

Experimental Design with One Independent Variable: Advantages and Disadvantages of Within- and Between-Subjects Designs

Power: a greater ability to detect significant findings.

It does not prove causation

Probably the most common limitation associated with correlational research is that

Has an equal chance of being selected

Random sampling means that each member of a particular population

Ratio

Sources of Research Ideas: Real Life Practical Problems Previous Research Theory

Real Life—Interested in investigating something you see in your own life? Sure! But before you plan your study, see what researchers already know about the topic. Practical Problems—Researchers sometimes seek a solution to a real-world problem. Again, before you plan your study, see what researchers already know about the topic. Previous Research—is often a source of research ideas. To learn about previous work: Read through abstracts of published research. An abstract is a short summary of a research report. Read a review article—provides a comprehensive summary regarding what's been done on a topic, but without including a full research report for any of the research projects it covers. Read a meta-analysis—a statistical procedure that combines the results of multiple studies on the same topic. Once you are familiar with what others have done on your topic of interest, you may choose to: Vary an independent variable differently Change the way you measure something Extend the external validity Theory—serves as a basis for formulating testable predictions about a phenomenon. So theories are often the source of research ideas. If the research conducted to test the theory provides data we expect, our confidence in that theory grows. If the research does not provide data we expect, modify the theory and test again (more ideas for research!).

People who take the Scholastic Aptitude Test and get an extremely low score, sometimes take the test twice. When they take the test a second time, they are likely to get a score slightly closer to the mean SAT score for the overall population of SAT test takers. What term describes what is happening here?

Regression toward the mean

Chapter 6 Correlation Methods and Statistics Magnitude, Scatterplots, and Types of Relationships

Relationships Between Variables

Test/retest

Repeated testing of the same individuals is used to determine _____ reliability.

results can be applied to subjects and settings other than those directly tested

Research possesses external validity if

True

Research published in journals usually undergoes the peer review process.

What is the APA's position with regard to confidentiality?

Researchers must protect participants' confidential information either by refraining from asking for identifying information or at least by keeping all identifying information separate from the research data.

Response bias refers to A habitual way of answering questions

Experimental Design with One Independent Variable: Analysis for One Independent Variable with Two Groups: A Sample Write-Up

Restaurant patrons who were given candy with their check left a larger tip percentage (M = 17.84, SD = 3.06) than restaurant patrons who were not given candy with their check (M = 15.06, SD = 1.89), t (90) = 5.25, p < .001, effect size r = .48. •The sentence is parallel. It makes the needed comparison between the two groups (those who were provided candy versus those who were not). •The mean (average) and standard deviation (an indicator of variability) of the responses for each condition are provided in parentheses, abbreviated and italicized, stated immediately after each condition is identified. •The details at the end of the sentence give information regarding the inferential statistic that was used. The "t" indicates that a t-test was used. The "(90)" represents the degrees of freedom for this test (in very general terms, this is a reflection of sample size). The "p < .001" represents the probability value that was calculated for this comparison. The effect size notation provides information regarding the magnitude of the result.

Case study method

SIgmund Freud built his theory of Psychoanalysis by using a detrailed record of a few patients of his. This is an example of which type of descriptive method?

Prediction

Scientists make predictions regarding their phenomenon of interest. These are called hypotheses. Hypotheses are our expectations for our results, and they are often developed from theories. Once we have formed our hypotheses, we can test them to determine how accurately they predict what occurs.

Explanation

Scientists often propose a theory to explain a phenomenon of interest. A theory is a statement that organizes, summarizes and explains available information about a phenomenon and serves as a basis for formulating testable predictions about the phenomenon.

Which of the following is not one of the ways that control is used in an experiment?

Scores on the dependent variable are controlled by the researcher

Stratified sampling

In my experiment I have 3 treatment conditions, labeled A, B and C. For each participant I present the treatment conditions in the following order: ABCCBA. Which type of counterbalancing have I used?

Subject by subject counterbalancing

double barreled question

Suppose my survey contains the question "Have we spent too much money on the war on drugs, and should Marijuana be legalized?" This type of question is known as a

In a single case design, when carry over effects are a concern, one way to deal with carry over effects would be to

Switch to a multiple baseline design

Repeated testing of the same individuals is used to determine _____ reliability.

Test/retest

Repeated testing of the same individuals to determine _____ reliability.

Test/retest

In which section of an APA style article would you find a complete summary of the research article?

The abstract

The title of an experiment is "The effects of candy on hyperactivity." The independent variable is

The amount of candy consumed by each participant

Internal validity refers to the extent that

The changes in the dependent variable can be attributed to changes in the independent variable

Which of the following is a necessary feature of a hypothesis?

The hypothesis should be stated after a study is conducted.

The title of an experiment is "The effects of candy on hyperactivity." The dependent variable is

The level of hyperactivity

Which of the following represents a possible operational definition of violent behavior in school-age children?

The number of times a child slaps or strikes another child

Two students replicated the Pepsi taste test. First, they gave volunteer subjects a taste of Coke. Next, they gave subjects a taste of Pepsi in unmarked glasses. This same ordering of drinks was presented to every subject. The majority of the subjects preferred Pepsi. What could you conclude about this research?

The order of the drinks was confounded with the independent variable

A quasi-experiment differs from a true experiment because in a quasi-experiment

The researcher selects instead of manipulates the levels of the independent variable

If a main effects is found in a factorial experiment, then

The scores made under one level of an independent variable were different from the scores made under another level of the same independent variable

dependent variable is The level of hyperactivity

The title of an experiment is "The effects of candy on hyperactivity." The dependent variable is

independent variable is The amount of candy consumed by each participant

The title of an experiment is "The effects of candy on hyperactivity." The independent variable is

Which of the following pieces of information is found in italics when writing a references section?

The title of the journal

Which of the following pieces of information is found in italics when writing a references section?

The volume number of the journal

Experimental Design with One Independent Variable: Measuring Dependent Variables

There are a variety of ways to measure dependent variables; they generally fall into three categories: •1) self report measures •2) behavioral measures •3) physiological measures •Self-Report Measures—Asking people to report what they think, what they feel, what they've done and what they think they would do in a given situation. •Self-reports can be responses to either open-ended or closed-ended questions. •Advantage: Self-reports can provide interesting insights. •Limitations: People may not tell the truth or even know the truth. In addition, sometimes being assessed can make people change their responses—this is known as reactivity. •Physiological Measures—there are many different types of physiological measures (such as heart rate, blood pressure). •Advantage: Physiological measures are less compromised by reactivity. •Limitation: Some physiological responses, such as brain activation need to be assessed with expensive equipment that may not be readily available to researchers.

A researcher finds an r = 0. She can conclude

There is no linear relationship between the variables

Festinger and his colleagues decided to join a group that believed the world was going to end. "Marian Keech," the leader of this group, said she was relaying these messages from planet Clarion. Keech and her followers believed they would be safe from this disaster because flying saucers would be arriving to rescue them. Festinger and his colleagues joined the group to see what Keech and her followers would do when the world didn't end. What form of research method were Festinger and his colleagues using and why?

They used participant observation because being undercover was the only way to be privy to the thoughts and actions of this group.

The Basic Components of a Journal Article

Title and author information Abstract Introduction Method Participants Apparatus/Materials Procedure Results Discussion References

Which of the following is not a goal of science?

To eliminate negative behavior

Why would a researcher use a ABA reversal design on a participant in a behavior modification study?

To make sure the treatment caused the change in behavior

Correlational research meets which goal of experimental psychology?

To predict behavior

A cross-sectional design uses different participants in each age group.

True

A factorial design must have at least 2 independent variables.

True

All other things being equal, the ANOVA run with a within-groups will have more statistical power than the ANOVA run with a between groups design.

True

If a significant interaction is found, it is followed up by a simple effects test, where we test the effect of one variable separately at different levels of the second IV.

True

In a 3 x 3 factorial design, it is possible to have 2 significant main effects but no significant interaction.

True

Is it possible to use a formula to compute a correlation that corrects for range restriction?

True

Research published in journals usually undergoes the peer review process.

True

he ABA reversal design would measure the participant in how many baseline conditions?

Two

In a graph of results, an interaction is shown by

Two lines that are not parallel to each other

If I reject my Null Hypothesis when it is actually true, I have made which error?

Type I error

If I FAIL reject my Null Hypothesis when it is actually FALSE, I have made which error?

Type II error

If I Fail to reject my Null Hypothesis when it is actually False, I have made which error?

Type II error

Which of the following situations best illustrates the principle of rationalism?

Using logic to deduce that "birds of a feather flock together."

In experiments, independent variables are

Varied by the researcher

alpha level) of .05, it means that I have a 5% chance of making a type I error

When I say that my study has a significance level (or alpha level) of .05, it means that

Nonlinearity:

When a correlation coefficient reveals that two variables are not correlated, it could mean that the relationship between them is not linear, but is curvilinear. Think Bell Curve | /\ | needs a different test

he researcher often goes undercover in the study

When collecting data using the participant observation method

Introduction

When you are writing a manuscript, where do you develop your rationale for doing your particular research project?

Which of the following would NOT be found in the Method section of an APA manuscript?

Whether they hypothesis was supported or not

an experiment taking place in a laboratory

Which method would usually involve debriefing participants after the study is over?

-0.94

Which of the following correlations indicates the strongest relationship between two variables?

To eliminate negative behavior

Which of the following is not a goal of science?

an experimental design

Which of the following methods would you use to determine a cause-and-effect relationship

The title of the journal

Which of the following pieces of information is found in italics when writing a references section?

The number of times a child slaps or strikes another child

Which of the following represents a possible operational definition of violent behavior in school-age children?

Naturalistic Observation

Which of the following research methods usually takes place in a natural setting?

Observing the friends you have to see if they tend to form relationships with people who are similar to them.

Which of the following situations best illustrates the principle of empiricism?

theory

Which of the following terms refers to a statement that organizes, summarizes, and explains available information about a phenomenon and serves as a basis for formulating testable predictions about the phenomenon?

A description of the statistical test used in the study

Which of the following would NOT be found in the Method section of an APA manuscript?

Introduction

Which of these sections comes first in an APA style manuscript?

Archival research

Which research technique would usually have the least amount of reactivity in particpants?

nonprobability sampling? Convenience

Which sampling technique is would be considered nonprobability sampling?

According to the APA ethical guidelines, it is acceptable to use deception in a study

Yes

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In this study of the new therapy, I reject my null hypothesis and conclude that the new therapy reduced depression symptoms, when in reality it has no effect on depression. I have made which type of error?

a type I error

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In this study of the new therapy, I fail to reject my null hypothesis and conclude that the new therapy has no effect on depression, when in reality it does lessen depression. I have now made which type of error?

a type II error

The best design to use for a single case (1 participant) experiment is

an ABA design

Which method would usually involve debriefing participants after the study is over?

an experiment taking place in a laboratory

Which of the following methods would you use to determine a cause-and-effect relationship

an experimental design

Following study participants over the course of many years can often become difficult. People sometimes cannot be tracked down and thus, they do not end up finishing the study. What is a term to describe this?

attrition

Reject the Null Hypothesis

break it down Failure means no Reject means no Null means none or same replace or write above it on the test Reject with Not and Null with Same so all this means is the Data collected did not "prove" (loosely applied here) that your original (idea/prediction/theory) was not right. in most cases that it was the opposite of your original (idea/prediction/theory) was right. don't you hate it, when that happens!

Failure to Reject the Null Hypothesis

break it down Failure means no Reject means no Null means none or same so just mark thru it on your test or cover it up on your screen or delete it from you mind so you are just left with "the null hypothesis" on the page which means the data collected agrees with what you predicted in your hypotheses

A curvilinear relationship is present when

changes in the value of one variable are accompanied by both increases and decreases in the other variable.

Of the two sources of error, which is more serious-extraneous variables or confounding variables?

confounding variables

If I were to test the same individuals at age 8, again at age 12, and again at age 16, I would be using which type of design?

cross-sectional

If I were to test three groups of individuals: some who were age 8, some who were age 12, and some who were age 16, I would be using which type of design?

cross-sectional

Suppose my survey contains the question "Have we spent too much money on the war on drugs, and should Marijuana be legalized?" This type of question is known as a

double barreled question

A successful replication with slightly different materials increases the __________________ of the findings.

external validity

When you can confidently state that your independent variable caused the difference observed in your dependent variable, your experiment is said to have

high internal validity

more on Regression to the mean

https://conjointly.com/kb/regression-to-the-mean/

In a factorial design, when the effect of an independent variable is different at different levels of the other independent variable, there is a(n)

interaction

QUESTION 3 If we can be certain that the new therapy reduced the number of depressive symptoms in the participants who were in the therapy group, we would say that the study has high

internal validity

If I were to test the same individuals at age 8, again at age 12, and again at age 16, I would be using which type of design?

longitudinal

Range restriction occurs when

most of the participants in a sample have similar scores for one of the measured variables in the correlation (the range of scores for one of the variables is restricted). This can reduce the strength of a correlation coefficient.

When collecting data using undisguised observation

participants are aware that they are being observed

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In the study of the new therapy, what would be the best way to decide who is in the experimental group and who is in the control group?

randomly assign 50 to the therapy group, and 50 to the control group.

In the study of the new therapy, what would be the best way to decide who is in the experimental group and who is in the control group?

randomly assign 50 to the therapy group, and 50 to the control group.

A goal of a within subjects design is to

reduce error between the groups introduced by individual differences between subjects

Research possesses external validity if

results can be applied to subjects and settings other than those directly tested

A researcher runs a correlational study and finds a Pearson's product moment correlation of r = 1.12. She can conclude

she must have calculated her correlation incorrectly

Applied research is done to

solve practical problems

strong positive correlation

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In the above study, what is the dependent variable?

the number of depressive symptoms in each participant at the end of the study

When collecting data using the participant observation method

the researcher often goes undercover in the study

How does a quasi-experiment differs from a true experimental design?

the researcher selects ,rather than manipulates the different levels of the IV

How does a quasi-experiment differs from a true experimental design?

the researcher selects,rather than manipulates the different levels of the IV

theory

A researcher runs a correlational study and finds a Pearson's product moment correlation of r = 0. She can conclude that

there is no linear relationship between the variables

What is the main purpose of counterbalancing in a within-subjects design?

to evenly distribute the progressive error across conditions

My new measure of intelligence actually measures reading comprehension instead of intelligence. We would say that this measure is low in

validity

Dependendant Variable (the thing Depends on Participants behavior)

what you are measuring ie. the number of depressive symptoms in each participant at the end of the study

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In this study, what is the independent variable?

whether the participant received the therapy or the control

An experimental study wishes to test a new type of therapy on depression symptoms, using 100 participants who have been diagnosed as suffering from Major Depressive Disorder. Fifty of the participants are assigned to the therapy group, and fifty are assigned to the control group. After 3 months, the number of depressive symptoms each participant is diagnosed with is measured. Use this scenerio to answer the following questions: In this study, what is the independent variable?

whether the participant received the therapy or the control

independent variable (the thing you are changing in the study.)

whether the participant received the treatment or (not)= the control

What type of design might use a Latin Squares to eliminate a potential confounding variable?

within-subjects design

Step 1: Develop a Research Idea: One Distinction to Consider

•1) Basic Research attempts to answer fundamental questions about a phenomenon without much focus on how the information could be applied to the real world. 2) Applied Research is conducted with a practical, real-world issue in mind.

Kuhn's Scientific Stages

•1. Preparadigmatic-Number of competing viewpoints exist. Results in random fact gathering (Psychology might be here) •2. Paradigmatic-single viewpoint dominates. This is normal science. •3. Revolutionary-Existing paradigm is being replaced by a new one. Occurs due to anomalies old paradigm can't account for.

Chapter 5 Descriptive Research Methods Case Studies

•A case study is usually an in-depth study of an individual, although sometimes it can be an in-depth study of an event or an organization. •Researchers in psychology typically use a case study to provide descriptive information about an unusual individual, perhaps someone who has a special ability or a disorder. •Advantage: •You generate a wealth of information about an unusual person. •Disadvantages: •Potential for observer bias. •No cause and effect. •Cannot generalize from your case study to other individuals.

( ch 6) Correlational Research Visually Representing Many Correlations

•A correlation matrix is used to present information about the correlations of many variables with each other. •Here is an example:

Small-N Designs

•A multiple-baseline design is a good choice of small-N design if you are studying behaviors that cannot be or should not be reversed, because in multiple-baseline designs, the treatment is never withdrawn. •There are three types of multiple-baseline designs: •1) multiple-baseline design across behaviors •2) multiple-baseline design across participants •3) multiple-baseline design across settings Here is an example of a multiple-baseline design across behaviors. This figure from Scattone (2008) illustrates the percentage of intervals containing three different types of social interactions.

Scales of Measurement •Nominal •Ordinal •Interval •Ratio •Your choice of scale determines the type of information you can describe given your data and what statistical analyses you can do with the data you generate.

•A nominal scale is the most basic form of measurement. It is characterized by the placement of information into categories. •If any numbers are associated with nominal scales, they do not have numerical meaning; you can only count the number of items placed into each category. •For example, the variable "political party" uses a nominal scale: Republican, Democrat and Independent are three common exemplars. •An ordinal scale is used when we want to rank-order the levels of the variable we are studying. •For example, Huffington Post ranked bacon (see www.huffingtonpost.com): •#1 (Best) Kirkland Signature Regular Sliced •#2 Oscar Meyer Thick Cut •#3 Great Value Lower Sodium Bacon (Walmart) •With the ordinal scale, we can't tell whether the intervals between the rankings are equal. •In an interval scale, there are equal intervals between the scale values. •Temperature uses an interval scale. The difference between 40 and 50 degrees is the same as the difference between 50 and 60 degrees. •You can perform standard arithmetic when you have an interval scale such as calculating an average temperature. •An interval scale does not have a true zero (there is no such thing as the absence of temperature). •A ratio scale has all the qualities of an interval scale (equal intervals between the scale values, and an ability can perform standard arithmetic) and a ratio scale also has a true zero. •Distance uses a ratio scale. The true zero means we can say that you have not gained any distance. A true zero also allows us to say, for example, that a distance of 4 miles is twice the length of 2 miles.

( ch 6) Correlational Research The Strength of a Correlation

•A number line has two anchor points or poles (-1.00 and +1.00). The closer a correlation coefficient is to its respective pole, the stronger the correlation. •A perfect correlation is the strongest correlation (either +1.00 or -1.00); it means that the two variables in question vary together perfectly. (see pic) •The numerical value in your correlation coefficient (the absolute value) tells you how strong your correlation is. •It is possible to calculate a probability value (p-value) to determine whether any given correlation is significantly different from zero. A probability value refers to the statistical likelihood that a result is due simply to chance. •Remember a correlation coefficient of zero means the two variables of interest are not correlated. •If you have a significant correlation this means you can predict a person's score on one measured variable when you know his or her score on the other measured variable. •The stronger the correlation, the better the prediction.

( ch 6) Correlational Research Visually Representing a Correlation

•A scatterplot (or scatterplot) is a graph of the relationship between two variables. •The horizontal axis (the x axis) indicates the possible scores for one of the variables. •The vertical axis (the y axis) indicates the possible scores for the other. •Each point on the scatterplot represents the intersection of an individual's scores on the two variables. •If you have a strong positive correlation, the plotted data will generally take the form of a diagonal line, going from the bottom left to the upper right of the graph (see "a" above). When the correlation is weaker, the diagonal will be less easy to visualize (see "b"). •If you have a strong negative correlation, you'll see a clear diagonal going from the bottom right to the upper left of the graph (see "e"). When the correlation is weaker, the diagonal will be less easy to visualize (see "d"). •The absence of a correlation is represented by the "c" graph.

Title Page

•A title page of a manuscript contains the title of the article and information about who did the research and where. •The Manual recommends that your title is no more than 12 words. •A common format for a title of a manuscript that describes an experiment is: Effects of Independent Variable(s) on Dependent Variable(s) •A title page of a manuscript contains the title of the article and information about who did the research and where. •The Manual recommends that your title is no more than 12 words. •A common format for a title of a manuscript that describes an experiment is: Effects of Independent Variable(s) on Dependent Variable(s) •The important thing to do when writing a title is "to summarize the main idea of a manuscript simply" (APA Manual, 2010, p. 23). •The title page of a manuscript also often includes an author note. This note includes information such as where to contact an author for more information.

Title Page

Operational Definitions

•A variable is any characteristic that can take on different values. •Every variable in a research study must be operationally defined. •To operationally define a variable means to specify the precise meaning of the variable in terms of the specific procedures to be performed. •The operational definition for an independent variable (IV) is the specific method used to manipulate the variable. •For example, Howell and Giuliano (2011) wanted to determine whether the use of expletives affected the effectiveness of sports coaches. Thus, they manipulated the presence/absence of expletives by sports coaches and operationally defined it in the following way: •Participants were randomly assigned to read a halftime speech in which a high school coach spoke with expletives or not. •The operational definition for a dependent variable is the specific method used to measure the variable. •For example, Gurung and Vespia (2007) wanted to know if a professor's appearance affected the professor's likeability. •How did they operationally define the dependent variable: the professor's likeability? Participants rated how much they like their professor (1 = do not like at all) (7 = like very much).

Step 2: Operationalize Your Variables

•A variable is any characteristic that can take on different values. •To operationalize your variables you would specify the precise meaning of a variable in terms of the specific procedures to be performed. •For example, a possible operational definition of aggression in a baseball game is the number of batters hit by the ball per game.

Chapter 6 Correlation Methods and Statistics Misinterpreting Correlations: Restrictive Range

•A variable that is truncated and has limited variability (Figure 6.3 Restricted range and correlation)

ADVANTAGES AND DISADVANGAGES OF INTERVIEWS (Survey Research)

•Advantages -Flexibility in collecting data -Interviewer can set tone and agenda •Disadvantages -Expensive -Lack anonymity, so responses might not be honest Lack standardized questions

EVALUATING SURVEY RESEARCH (Survey Research)

•Advantages -Permits good generalization -Efficient data collection -Can yield very accurate results •Disadvantages -Bias •Interviewer bias •Interviewee bias -Non response

Step 7: Analyze the Results of Your Study

•After you have collected your data, it's time to analyze your results. •Descriptive statistics—summarize the data. •Inferential statistics—determine whether the mean responses for the groups were significantly different from each other. Inferential statistics theoretically enable researchers to make conclusions about a population after studying just a portion of that population.

Explanatory Method

•Allows a researcher to establish a cause-and-effect relationship -Through manipulation of a variable and -Control of the situation •Independent variable: variable that is manipulated •Dependent variable: variable that is measured •Random assignment: assigning subjects to conditions in such a way that every participant has an equal probability of being placed in any condition •Control: manipulating the independent variable or any other extraneous variables that could affect the results

Experimental Design with One Independent Variable: Manipulating Independent Variables

•Amount of a Variable •For example:

Experimental Design with One Independent Variable: Instructional Versus Event Manipulations

•Another possibility to consider is whether to use an instructional versus an event manipulation. •An instructional manipulation just alters the instructions presented to study participants while an event manipulation alters the experiences that participants encounter. •An event manipulation tend to be higher in experimental realism—the degree to which a research experience psychologically involves the experimental participants.

Experimental Design with One Independent Variable: Individual Difference Manipulation

•Another way to vary an independent variable is to do an individual difference manipulation. In this case the researcher is putting people into groups based on some personal characteristic that the individual has (such as gender, religion). •Since groups are not considered equivalent (people were not put into the different groups randomly), you can't state cause and effect with certainty.

Conferences

•Attend a professional conference! It's a great way to learn about current research! •Conferences typically provide two ways to present research: •1) researchers present their work on a poster, •2) researchers speak to an audience •In a conference poster session, presenters and their posters line a room, usually for an hour or two, while an audience circulates among the posters. The researchers are there to discuss their work and answer questions. •Those who are speaking to an audience usually have a visual component to their talk (such as PowerPoint slides).Talks are usually about 10-20 minutes long and are typically grouped with others on similar topics. Researchers usually answer questions at the end of the session.

Basic and Applied Research

•Basic research -Seek knowledge for its own sake -May be used to test theories or hypothesis -Example: identifying whether cognitive maps can be mentally rotated •Applied research -Practical significance and potential solutions -Example: identifying factors associated with weight gain

Step 6: Pilot Test and Then Conduct Your Study

•Before you conduct your study, you need to conduct a pilot test—a series of practice sessions run during the initial stages of research that allow us to determine whether the procedure is running as intended. •Once you've pilot tested your study, you are ready to collect data!

Experimental Design with One Independent Variable: The Posttest Only Design Versus the Before-After Design

•Before-After Design (also known as the pretest-posttest design): the dependent variables are measured both before and after the independent variable is manipulated: •After manipulation of the independent variable, researchers would analyze how much the groups changed from pretest to posttest; this would allow them to determine which group showed the greater amount of change.

Chapter 6 Correlation Methods and Statistics Misinterpreting Correlations: The Assumptions of Causality and Directionality

•Causality: the assumption that a correlation indicates a causal relationship between the two variables •Directionality: the inference made with respect to the direction of a causal relationship between two variables •For example, does watching violent TV lead to aggression, or does aggression lead to preference for violent TV programs?

TRUE EXPERIMENTAL RESEARCH METHODS

•Chapter 11 will cover quasi-experimental designs. -Quasi-experimental designs involve independent variables that are not under the direct control of the experimenter, but instead, occur naturally. -Variables such as gender, political affiliation, and mental health status are quasi-variables. •By contrast, true-experiments involve manipulating an independent variable. -True experiments allow stronger statements about causality than do quasi-experiments. -This presentation will help you to compare and contrast the design ideas in this chapter to those in chapter 11. •Allow statements about cause and effect -By controlling potential sources of variance •The simplest experimental design -Randomly selects subjects from population -Experimental group—receives treatment -Control group—does not receive treatment •Characteristics -Random assignment -Control group •Three typical designs -Pretest post-test control group design -Post-test only control group design -Solomon four-group design

SOLOMON FOUR GROUP DESIGN

•Consider the teaching math example -If you compare the post-test scores of experimental and control group 1 to those of control groups 2 & 3, you will assess the effects of pre-test vs. no pre-test without regard to the methods of teaching. -If you compare the post-test scores of experimental and control group 2 to those of control groups 1 & 4, you will assess the effects of the method of teaching without regard to the pretest. If you compare the post-test scores of all four groups to each other, you will be able to determine whether the combined effects of the pre-test and method of teaching are different from the effects of either alone. •Most researchers consider the Solomon Four Group Design to be the best in terms of internal validity. •But it is expensive and time consuming to use.

Characteristics of Good Measurement: Validity •Validity of a measure is the degree to which it measures what you intend it to measure. •Validity can be divided into two general categories: construct validity and criterion validity. •

•Construct Validity is the degree to which a test actually measures the construct it is designed to measure. There are many ways to assess construct validity: •Face validity •Content validity •Convergent validity •Discriminant validity •Face validity indicates how well a measure, on the face of it, appears to measure what it was designed to measure. •Content validity is the degree to which the items on the test adequately reflect all components of the construct you are measuring. •Convergent validity is the degree to which a measure is related to other tests designed to measure the same general construct. •Discriminant validity (divergent validity) is the degree to which a measure is not associated with measures of different constructs. •Criterion validity refers to the idea that the measure in question is related to a specific relevant outcome. •There are two types of criterion validity: predictive validity and concurrent validity. •Predictive validity tells us whether a test can accurately predict a particular future behavior. •Concurrent validity is present when scores on a test correlate with an already validity measure given at the same time (concurrently). •Internal validity asks the question: To what extent can I conclude that my independent variable caused my dependent variable responses? •When you can confidently state that your independent variable caused the difference observed in your dependent variable, your experiment is said to have high interval validity (only relevant for experimentation). •External validity is the extent to which you can generalize your results to other persons, places and times (relevant to both experimental and non-experimental research).

NONPROBABILITY SAMPLING STRATEGIES (Survey Research)

•Convenience sampling -Captive or easily sampled population •E.g., the students in a class, or people coming out of a theatre. -Not random -Weak representativeness •Quota sampling -Proportional stratified sampling is desired but not possible -Participants with the characteristic of interest are non-randomly selected until a set quota is met

( ch 6) Correlational Research The Direction of a Correlation

•Correlation coefficients are either positive or negative. •A positive correlation between two variables means those variables tend to move in the same direction. Higher scores on one variable tend to be associated with higher scores on the other variable. Lower scores on one variable tend to be associated with lower scores on the other variable. A negative correlation between two variables means those variables tend to move in opposite directions. Higher scores on one variable tend to be associated with lower scores on the other variable

Predictive (Relational) Methods

•Correlational method: assesses the degree of relationship between two variables •Correlation does not imply causation •Positive relationship: an increase in one variable is accompanied by an increase in the other variable •Negative relationship: an increase in one variable is accompanied by a decrease in the other variable •Quasi-experimental method: compares naturally occurring groups of individuals; the variable of interest cannot be manipulated •Subject (participant) variable: a characteristic inherent in subjects that cannot be changed

( ch 6) Correlational Research

•Correlational research is a non-experimental research approach in which variables are measured, not manipulated. •In correlational research, we measure different variables to see whether there is a relationship between or among them. •We typically represent a correlational relationship with a correlation coefficient, a numerical value that ranges from -1.00 to +1.00. Picture a number line: ------------------------------------------------- -1.00 0 +1.00 •Most correlation coefficients are the result of an analysis called a Pearson product-moment analysis (Pearson r). •A Pearson r is a statistical analysis that identifies the direction and strength of the linear relationship between two variables on an interval or ratio scale. •A linear relationship is a relationship that can be approximated with a straight line on a graph.

Chapter 6 Correlation Methods and Statistics Summary

•Correlations vary in type and magnitude •Errors are commonly made when interpreting correlations •Correlation coefficients: Pearson's, Spearman's, Kendall's •Coefficient of determination and regression analysis provide a tool for predicting from one variable to another

(Survey Research ch 7) Survey Research: Types of Questions

•Demographic questions are designed to assess the characteristics of the respondents (such as age, gender, education level). •Open-ended questions—a type of question that does not have fixed answers but instead allows the respondent to answer in any way. •Advantage: respondents may give you information you had not considered before. •Disadvantage: you may get answers that are not helpful. •Need a content analysis to analyze the response from open-ended questions. Read through all the responses and then create categories that will allow you to summarize the information. •Another type of question is a closed-ended question (also known as a restricted question or a fixed alternative question). With this type of question people are asked to respond by choosing one or more answers from a set of offered options. •Advantage: It is much easier to summarize the results of a closed-ended than a open-ended question. Just report the percentage that chose each option. •One commonly used closed-ended question format is the Likert scale which provides study participants a way to indicate their level of agreement or disagreement with a statement. For example: |------------------|-----------------|------------------|-----------------| 1 2 3 4 5 Strongly Disagree Neither Agree Agree Strongly Disagree nor Disagree Agree •Another type of closed ended question format is the semantic differential scale which provide respondents with response options that are paired adjectives that are opposites of each other. For example: Binge Drinking Unenjoyable _____ _____ _____ _____ _____ Enjoyable •You can combine open- and closed-ended questions in the same questionnaire. •When you provide open-ended questions followed by closed-ended questions, it is called funneling (broader questions before specific questions).

DEVELOPING AN INTERVIEW (Survey Research)

•Describe goals of project •Select appropriate sample •Develop interview questions •Train interviewers •Conduct interviews

Chapter 6 Correlation Methods and Statistics Conducting Correlational Research

•Determine whether two variables are related to each other -Example: do people who are taller tend to weigh more? •Allow us to make predictions from one variable to another -Example: estimate an individual's weight based on height •Reasons -Ethically impossible to do an experimental study -Useful in measuring many variables and assessing the relationships between them

Determinism

•Determinism is a very important concept in Experimental Psychology •The cause of any event can be determined, because it has finite causes •For everything that ever happens there are conditions such that, given them, nothing else could have happened •Leaves no place in behavior for free will

HOW TO DO SURVEY RESEARCH (Survey Research)

•Develop a plan 1.Clarify the objectives 2.Identify a sample 3.Define a method •How will the questions be structured? •How will the data be collected? •What types of questions will be asked? 4.Coding and scoring What do the numbers mean?

The Steps in the Research Process

•Develop a research idea •2) Operationalize your variables •3) Choose a research approach •4) Form hypotheses •5) Recruit study participants •6) Pilot-test and conduct your study •7) Analyze the results of your study •8) Report what you did •9) Start the process all over again

Clarify the Objectives (Survey Research)

•Do you want to— -Assess political attitudes or predict voting patterns? -Gather data to help market a new product? -Find out what students think about a proposed academic program? -Determine whether child abuse is common among a particular segment of the population? -Assess whether a community will support a new mental health facility?

BASIC ASSUMPTIONS OF A QUESTIONNAIRE (Survey Research)

•Does not make unreasonable demands upon the respondent •Does not have a hidden purpose •Requests information that respondents presumably have •Contains interesting questions •Does not request information that could be obtained by other means

PRETEST POST-TEST CONTROL GROUP DESIGN

•E.g., Hypothesis—A new method of teaching math will help students learn better than the old method. -Randomly assign students to new vs. old method -Pre-test math ability -Teach with old or new method Assess results

References

•Every paper has a Reference section in which you place citations for all the references you cited in your paper. •All references are listed in alphabetical order (use the last name of the first author to decide where the source belongs in your alpha list). •The format of all references is specified by the APA Publication Manual. •For example, here is the general format for a journal article with one author: Last name, Initial of first name. Initial of middle name. (Date of publication). Title of journal article. Name of Journal, Volume(Number), pages. doi:xxxxxxxxx •You may have noticed that the reference citation shown above has a digital object identifier (DOI). This DOI is a unique sequence of letters and numbers that can be used to find an article in a search at CrossRef.org.

Chapter 6 Correlation Methods and Statistics Prediction and Correlation

•Example: -Strong positive correlation between smoking and cancer -You may know someone who has smoked for 30 years and does not have cancer •Person-who argument -Arguing that a well-established statistical trend is invalid because we know a "person who" went against the trend

Experimental Design with One Independent Variable

•Experimental design is the plan or strategy you use when conducting an experiment. •When conducting the simplest kind of experiment, researchers manipulate (vary) one variable, called the independent variable, and observe the effects of that manipulation on a response measure called the dependent variable. •The independent variable is considered to be the cause and the dependent variable is the effect. •To establish cause and effect between an independent and a dependent variable, your experiment needs to fulfill two requirements: •1) it must have theoretically equivalent groups •2) it must have at least two groups of study participants.

Experimental Design with One Independent Variable: Order Effects in Within-Subjects Designs

•Fatigue effect—performance deteriorates as the participant becomes tired, bored, or distracted. •Practice effect—performance improves as the participant benefits from practice. •Carryover effect—performance in one condition affects performance in another condition. •Sensitization effect—participants can more readily determine the purpose of your study and/or your hypothesis since they are privy to all levels of your independent variable. •Practice, fatigue, carryover and sensitization effects are all called order effects and they need to be controlled. •One way to control these is to change the order of presentation of your conditions—this is called counterbalancing. •When you have a within-subjects design, counterbalancing is vital.

(Survey Research ch 7)Administering the Survey

•First, pilot your survey (do a practice run). •Ways to administer the survey: 1) Face-to-face interview •Advantages: •Can develop rapport •Can often judge immediately whether the respondent understands the question •Disadvantages: •Time consuming and expensive •Interviewer can actually bias the results •People might give different answers in a face-to-face interview 2) Telephone survey: •Advantages: •Cost effective •Can develop rapport •Disadvantages: •Telephone interviewing may be falling out of favor 3) Internet survey: •Advantage: •Cost effective •Disadvantages: •Drop-out rate tends to be relatively high •Possible non-response bias 4) Mail survey: •Advantage: •Cost effective •Disadvantage: •Possible non-response bias

PRETEST POST-TEST CONTROL GROUP DESIGN

•Groups should be equivalent at beginning SO •Observed differences must result from treatment •E.g., Hypothesis—A new method of teaching math will help students learn better than the old method. -Randomly assign students to new vs. old method -Pre-test math ability -Teach with old or new method -Assess results

Deceprion/ lieing

•Handle deception appropriately: •Deception must be "justified by the study's significant prospective scientific, educational or applied value." Only use deception if "effective nondeceptive alternative procedures are not feasible." •"Do not deceive prospective participants about research reasonably expected to cause pain or distress." •"Explain deception... as early as feasible." •"Permit participants to withdraw their data." •Various kinds of deception. Examples: •Cover story •Use of a confederate (accomplice of the experimenter) •The APA Ethics Code has guidelines for humane animal care and use in research. •Treat research animals humanely. For example, researchers are expected to "minimize the discomfort, infection, illness and pain of animal subjects." •Reporting research results: •"Psychologists do not fabricate data." "If psychologists discover significant errors in their published data, they take reasonable steps to correct such errors." •Avoid plagiarism: •"Psychologists do not present portions of another's work of data as their own." •Give publication credit: •"Psychologists take responsibility and credit...only for work they have actually performed." •Avoid duplicate publication of data: •"Psychologists do not publish, as original data, data that have been previously published."

Experimental Design with One Independent Variable: A Consideration of the Results

•How do we think about results from experiments with one independent variable? •One of the first steps that researchers complete after collecting data is to calculate descriptive statistics to summarize the data they collected from each group (for example, researchers will typically calculate the mean responses for each group). •Once descriptive statistics have been calculated, researchers often use inferential statistics to determine whether the mean responses for the groups were significantly different from each other. •In other words, we are asking if the results are likely due to the effects of the independent variable or likely due to chance.

STRUCTURED QUESTIONS (Survey Research)

•How frequently do you attend religious services per year? -None -1-10 times -11-25 times -26-50 times -More than 50 times •At what age do you expect to retire? -Younger than age 60 -Age 60-65 -Age 66-70 -71-75 years old Older than age 75

(Survey Research ch 7) Sample Size

•How large should your sample be? You want the sample to be large enough to represent the population adequately so you need to have a sense of how much diversity there is in your population. •If your population is homogeneous (all members are very similar to each other), your sample doesn't need to be large to capture a good range of responses. •If your population is heterogeneous (your population has variability), then your sample needs to be larger to represent that population adequately. •Larger samples generally represent populations better than smaller samples. •When thinking about sample size we also need to consider how much error we are willing to tolerate in our results. •A margin of sampling error acknowledges that there may be some measurement error when you look at only a sample of the population as opposed to the whole population.

The Research Process

•Identify a problem •Review the literature •Generate hypotheses •Design and conduct the study •Analyze the data and interpret the results •Communicate the results

PARTIALLY STRUCTURED QUESTIONS (Survey Research)

•If you eat at fast food restaurants, what type of fast food do you prefer? -Hamburgers -Chicken -Fish -Mexican -Other (specify)___ •What is your political affiliation? -Republican -Democrat -Libertarian -Green -Other (specify)___ -No affiliation

Experimental Design with One Independent Variable: Between-Subjects Designs

•In a between-subjects design, we randomly assign people to different groups. Groups are then theoretically considered to be equivalent. •Sometimes researchers will match participants to each other on some variable and then randomly assign the matched participants to groups. This is called a matched-groups design. Groups in this case are also considered to be equivalent.

Small-N Designs

•In a multiple-baseline design across participants, researchers provide a treatment for different individuals who have the same target behavior change as a goal. •The researchers provide the intervention to one participant at a time. If the individual exposed to the treatment changes while those who have not yet undergone the intervention remain at baseline, this provides evidence for the effectiveness of the treatment (see next slide). Luke and Alavosius (2011) used a multiple-baseline design across participants to improve adherence to hand hygiene rules in three medical staff members. This figure illustrates the mean percentage of hand hygiene components performed correctly over the course of successive work weeks. •Note that the multiple-baseline design across participants has a built-in test of the generalizability of treatment effects. The knowledge that a particular treatment worked to change behavior in different individuals increases your confidence in the generalizabilty of the findings.

Experimental Design with One Independent Variable: Extraneous Variables/Confounds

•In a well-designed experiment, the only difference between your groups is the independent variable. •Extraneous variables are all the uncontrolled variables that are not purposely manipulated but that could potentially affect our groups. Important to keep extraneous variables constant in an experiment. •The worst kind of extraneous variable is a confound—it varies systematically with the independent variable. •When a confound is present, you are not able to establish cause and effect. Confounds decrease your internal validity, the extent to which you can determine that your independent variable caused your results. •The worst kind of extraneous variable is a confound—it varies systematically with the independent variable. •When a confound is present, you are not able to establish cause and effect. Confounds decrease your internal validity, the extent to which you can determine that your independent variable caused your results.

Experimental Design with One Independent Variable: Within-Subjects Designs

•In a within-subjects design, we test the same people for different levels of our independent variable. This design is called a within-subjects design or a repeated-measures design. •Since all participants experience each level of the independent variable in a within-subjects design, it is important that you control the order in which you present the levels.

Chapter 5 Descriptive Research Methods Naturalistic Observation

•In naturalistic observation you observe people (or animals) in their natural settings and systematically record their behavior. •When conducting naturalistic observation, follow systematic procedures for recording observations. •1) Have clear definitions of the behaviors of interest. This can help you avoid observer bias which can occur when the observer has a preconceived notion of how those being observed are likely to act. •2) Use a coding system so you can categorize your data for later analysis. •When conducting naturalistic observation, follow systematic procedures for recording observations. •1) Have clear definitions of the behaviors of interest. This can help you avoid observer bias which can occur when the observer has a preconceived notion of how those being observed are likely to act. •2) Use a coding system so you can categorize your data for later analysis. •3) Make reliable (consistent) observations. Check for inter-rater reliability. •4) Be unobtrusive when observing. If participants know they are being observed, that knowledge may affect their behavior. This is called reactivity. •One option is to have those you are watching habituate (get used to) to your presence. •Advantage: •Since you are observing people in their natural environments, you typically can have confidence that you are seeing true behavior. •Disadvantages: •Cannot establish cause and effect. •Ethical issues (informed consent and debriefing are typically not a part of the process).

Chapter 5 Descriptive Research Methods Participant Observation

•In participant observation, the researcher observes and participates in the observed activities, in some cases interacting with the study participants. •The researcher is often "undercover" when conducting a participant observation study. •When conducting participant observation, follow systematic procedures for recording observations. •1) Have clear definitions of the behaviors of interest. This can help you avoid observer bias. •2) Try not to influence those you are watching. •3) Record observations secretly. •Advantage: •Again you typically can have confidence that you are seeing true behavior. •Disadvantages: •Cannot establish cause and effect. •Ethical issues (informed consent and debriefing are typically not a part of the process).

Results

•In the Results section you will give your readers information about your statistical analyses. •If the research was an experiment, you will reveal how different groups performed relevant to each other. •If the research was not an experiment, you should describe how the participants performed overall. •If you conducted an experiment, begin your Results section with information about any manipulation checks you included in your work. •Tie your results to your hypotheses so readers can easily see whether your hypotheses were supported. •Organize your results so that they are easy to understand (for example, report on one dependent variable at a time). Start with simpler results and move toward more complex results (such as main effects, followed by interactions).

( ch 6) Correlational Research Correlations

•In the middle of the number line is a zero that represents the absence of a correlation (neither positive nor negative).

Small-N Designs

•In the multiple-baseline designs across settings, researchers provide treatments for one individual in multiple settings, targeting one setting at a time. Dalton et al. (1999) used a multiple-baseline design across settings to reduce off-task behavior across three settings in two adolescents, each shown in the figure on the right

POPULATIONS AND SAMPLES (Survey Research)

•Inferential method is based on inferring from a sample to a population •Sample—a representative subset of the population •Population—the entire set of participants of interest •Generalizability—the ability to infer population characteristics based on the sample

USE A COVER LETTER WHEN SENDING OUT QUESTIONNAIRES (Survey Research)

•Informs the recipient about the research •Establishes the importance of the research •Makes the recipient a part of the research

CLUSTER SAMPLING (Survey Research)

•Instead of randomly selecting individuals -Units (groups) of individuals are identified •E.g., high schools or voting precincts -A random sample of units is then selected •E.g., Lincoln, Washington, & Meriweather high schools. •10 of 40 voting precincts. -All individuals in each unit are assigned to one of the treatment conditions •Units must be homogeneous in order to avoid bias

INTERVIEWS ARE THE BASIC TOOL OF SURVEY RESEARCH (Survey Research)

•Interviews allow flexibility in collecting data •The interviewer can set the tone -Can change agenda for interview -Can ask follow-up questions

SCORING LIKERT RESPONSES Method of Summated Ratings (Survey Research)

•Items are weighted •Weights of unfavorable items are reversed •Average score is computed Item 1 = 5 Item 2 = 1 Item 3 = 4 Item 4 = 2 Item 5 = 4 Items 1 & 3 are reversed Item 1 = 1 Item 3 = 2 Score = (1 + 1 + 1 + 2 + 4)/ 5 Attitude = 1.8

How do you make your groups equivalent? Experimental Design with One Independent Variable

•Let's say you have one independent variable with two groups. You can: •1) Randomly assign your study participants to the two groups—this is a between-subjects design. It has different people in the different groups. •2) Test the same people in both groups (test your participants twice)—this is a within-subjects design. It has the same people in the different groups.

Chapter 6 Correlation Methods and Statistics Magnitude, Scatterplots, and Types of Relationships

•Magnitude: an indication of the strength of the relationship between two variables •0 - .29 weak relationship •.30 - .69 moderate relationship •.70 - 1.00 strong relationship •Remember, the sign only tells us what type of relationship we have. •Scatterplot: a figure that graphically represents the relationship between two variables ( fig 6.1 Scatterplot for height and weight)

Chapter 6 Correlation Methods and Statistics Magnitude, Scatterplots, and Types of Relationships

Experimental Design with One Independent Variable: Which Dependent Variables Should You Choose?

•Most experiments include more than one dependent variable, and you can also use more than one type of dependent variable (self-report, behavioral, physiological). •Choose the variables that you expect to be sensitive to your independent variable manipulation.

Chapter 5 Descriptive Research Methods Types of Descriptive Research Methods

•Naturalistic Observation •Participant Observation •Archival Research •Case Studies

Descriptive Research Approaches

•Naturalistic observation—we observe people or animals in their natural settings and systematically record their behavior. •Correlational research—we measure two or more variables to see whether there is a relationship between or among them. •Survey—we attempt to estimate the opinions, characteristics, and/or behaviors of a particular population by asking some of the people in that population to respond to questions.

External Validity

•No matter what kind of research approach you use, the question—"Do these results hold for others?" is relevant. This is a question of external validity-the generalizability of the results to other persons, places, or times.

Experimental Design with One Independent Variable: Avoid Experimenter Expectancies

•No matter what type of dependent variable you use, be careful to avoid the influence of experimenter expectancies. •Experimenter expectancies refers to the influence of an experimenter's expectations on the results of an experiment. •How to guard against the influence of experimenter expectancies? Use a double-blind experiment—one in which the participants and the researchers interacting with them do not know the group assignments or the hypotheses.

( ch 6) Correlational Research Factors Affecting the Strength of a Correlation Coefficient

•Nonlinearity: When a correlation coefficient reveals that two variables are not correlated, it could mean that the relationship between them is not linear, but is curvilinear. •A curvilinear relationship is present when changes in the value of one variable are accompanied by both increases and decreases in the other variable. •Range restriction occurs when most of the participants in a sample have similar scores for one of the measured variables in the correlation (the range of scores for one of the variables is restricted). This can reduce the strength of a correlation coefficient. •It is possible to use a formula to compute a correlation that corrects for range restriction.

Hypothesis Testing: Null and Alternative Hypotheses

•Null hypothesis (Two-tailed test) -The hypothesis predicting that no difference exists between the groups being compared -Statistical notation: H 0: μ1 = μ2 •Alternative hypothesis (Two tailed test) -The hypothesis that the researcher wants to support, predicting that a significant difference exists between the groups being compared -Statistical notation: H 1: μ1 ≠ μ2

Descriptive Methods

•Observational method: making observations of human or animal behavior •Naturalistic observation: observing behavior of humans or animals in their natural habitat •Laboratory observation: observing behavior in a more controlled situation •Case study method: in-depth study of one or more individuals •Survey method: questioning individuals on a topic and then describing their responses •Sample: group who participate in a study •Population: all of the people about whom a study is meant to generalize •Random sample: achieved through random selection in which each member of the population is equally likely to be chosen

Step 4: Form Hypotheses

•Once you have selected your research idea, operationalized your variables, and selected your research approach, you are ready to form your hypotheses, your predictions for the outcomes of your research.

Hypothesis Testing: One- and Two-Tailed Hypothesis Tests

•One-tailed hypothesis (directional hypothesis) -An alternative hypothesis in which the researcher predicts the direction of the expected difference between the groups -Null hypothesis: H 0: μ1 ≤ μ2 Alternative Hypothesis: H 1: μ1 > μ2 •Two-tailed hypothesis (nondirectional hypothesis) -An alternative hypothesis in which the researcher predicts that the groups being compared differ but does not predict the direction of the difference -Most of the time we use this type of hypothesis in research

Chapter 6 Correlation Methods and Statistics Prediction and Correlation: Correlation Coefficients

•Pearson product-moment correlation coefficient (Pearson's r) -The most commonly used correlation coefficient when both variables are measured on an interval or ratio scale -This is the correlation we have been discussing so far. •Spearman's rho and Kendall's tau can be used with ordinal data -They are interpreted the same, though. •Coefficient of determination (r2) -A measure of the proportion of the variance in one variable that is accounted for by another variable -Calculated by squaring the correlation coefficient -This gives an estimate of how much one variable influences the other variable (like effect size)

Experimental Design with One Independent Variable: The Posttest Only Design Versus the Before-After Design

•Posttest Only Experiments: the dependent variables are measured only after the independent variable has been manipulated and not before: •After manipulation of the independent variable, researchers would analyze statistically how much the groups differed from one another.

General Principles of the American Psychological Association's (APA) Ethics Code •Influenced by both the Nuremberg Code and the Belmont Report, the APA first published its Ethics Code in 1953. •The APA Ethics Code contains 5 aspirational principles (not obligatory): •Principle A: Beneficence and Nonmaleficence •Principle B: Fidelity and Responsibility •Principle C: Integrity •Principle D: Justice •Principle E: Respect for People's Rights and Dignity

•Principle A: Beneficence and Nonmaleficence—"Psychologists strive to benefit those with whom they work and take care to do no harm." •Principle B: Fidelity and Responsibility—"Psychologists should establish relationships of trust with those with whom they work." •Principle C: Integrity—"Psychologists seek to promote accuracy, honesty, and truthfulness in the science, teaching, and practice of psychology." •Principle D: Justice—"Psychologists recognize that fairness and justice entitle all persons to assess to and benefit from the contributions of psychology." •Principle E: Respect for People's Rights and Dignity—"Psychologists respect the dignity and worth of all people, and the rights of individuals to privacy, confidentiality, and self-determination." •The APA Ethics Code also contains Ethical Standards that are mandatory rules of conduct for all student affiliates and members of the APA. Here are the major rules related to research. •Obtain approval from an Institutional Review Board (IRB) before conducting your research. •Avoid harming your clients/patients, students, research participants. •Protect confidential information. •Obtain informed consent. In almost all cases, psychological researchers must provide enough information about the study in order for participants to be able to make an informed decision about participating. •Usually this entails reading over and signing an "informed consent form." The APA Ethics Code indicates what is to be placed on this form. •Respect client/patient, student and subordinate research participants: •No "adverse consequences of declining or withdrawing from participation." •"When research participation is a course requirement or an opportunity for extra credit, the prospective participant is given the choice of equitable alternative activities." •Offer reasonable inducements for research participation.

(Survey Research ch 7) Probability Sampling

•Probability sampling: each study participant has a known probability of being included in the sample. •1) Simple random sampling: Select a random sample of respondents—every person in the entire population of interest has an equal chance of being selected. •If you select your participants randomly from the population, your sample will be representative. A representative sample is one that has the same relevant characteristics as the population of interest. •If you sample randomly from your target population, in theory you can infer that your obtained results hold for the rest of the target population. •2) With the probability sampling technique called stratified sampling, you divide your population of interest into segments called strata. Each stratum represents a group of people who have one or more characteristics of interest. •Once you divide your target population of interest into strata, then you choose a random sample from each of the segments. •3) Cluster sampling is another probability sampling technique. The researcher identifies clusters of individuals then randomly samples the clusters to decide which ones to use. All participants within the chosen clusters are surveyed.

CHOOSING A REPRESENTATIVE SAMPLE (Survey Research)

•Probability sampling—the likelihood of any member of the population being selected is known •Nonprobability sampling--the likelihood of any member of the population being selected is unknown

Chapter 6 Correlation Methods and Statistics Prediction and Correlation: Advanced Correlational Techniques: Regression Analysis

•Regression analysis -A procedure that allows us to predict an individual's score on one variable based on knowing one or more other variables •Regression line -The best-fitting straight line drawn through the center of a scatterplot -Indicates the relationship between the variables height and weight for this group of individuals ( Fig 6.4 The relationship between height and weight with the regression line indicated)

Characteristics of Good Measurement: Reliability

•Reliability—the ability of a measure to produce consistent results every time. •When variation in the measurements is present, we say there is measurement error. There are two kinds of measurement error. •Systematic error—a constant amount of error each time a measurement is taken. •Random error—a nonsystematic fluctuation in the measurement. •Test-Retest Reliability—Administer the same test to the same people at two different times under equivalent test conditions. Results should be very similar each time. •To determine how well the scores at the two testing sessions match, we calculate a correlation coefficient (Pearson r) which is a numerical value that ranges from -1.00 to +1.00. •The correlation coefficient provides information about both the strength and the direction of the relationship between the two scores we are measuring. •A correlation coefficient ranges from -1.00-+1.00. Picture a number line: ------------------------------------------------------------- -1.00 0 +1.00 •Strength of a correlation: The closer a correlation coefficient is to either -1.00 or +1.00, the stronger it is. •Direction of a correlation: Direction is indicated by the sign of the correlation, either positive or negative. •A positive correlation means the scores move in the same direction (for example, when the scores are high on the first variable, they are high on the second variable). •A negative correlation: means the scores move in opposite directions (for example, when the scores are high on the first variable, they are low on the second variable). So when we are assessing test-retest reliability, ideally, we want to see a strong, positive correlation •We can also assess internal consistency reliability—check to see if all the items on a test are related (measuring the same underlying attribute). •To assess internal consistency, you can check for split-half reliability. Divide the test items in half and calculate a Spearman-Brown split-half reliability coefficient. This is a correlation between the total score on one half of the test with the total on the other half. •Also possible to calculate Cronbach's α (alpha) coefficient which calculates all possible split-half correlations and averages them together. •We can also assess internal consistency reliability by computing item-to-total correlations. To do this, calculate the correlations between the score on each of the individual items and the total score (excluding the score for the item you are checking). •Eliminate any items that are not strongly correlated with the total score. •We can also assess inter-rater reliability when respondents answer open-ended questions. Think of open-ended questions as akin to essay questions—there are no response options to choose from. •When analyzing responses from open-ended questions, you need to perform a content analysis—read through all responses and then create categories that will allow you to adequately summarize the content. •Typically more than one researcher makes decisions regarding how to categorize open-ended responses, and you want those decisions to be consistent with each other. •Inter-rater reliability is the level of internal consistency among the judges. •Researchers use a statistic known as Cohen's Kappa (κ) to indicate the level of agreement among the judges. Kappa ranges from 1, which means the judges do not agree at all to +1.00 which means the judges are in perfect agreement. •Test-Retest reliability is a test of stability over time. •Split-half, Cronbach's α, and item-to-total correlations test for equivalence among the items on a test. •Inter-rater reliability is a test of equivalence between those making judgments about test responses.

Experimental Design with One Independent Variable: Manipulation Checks

•Researchers often use a special kind of dependent variable called a manipulation check. •A manipulation check is a dependent variable that determines whether your manipulation worked. For example, if you were manipulating the emotions induced in participants, you should check to determine whether the intended emotions were indeed induced (can ask participants how they feel). •If your manipulation didn't work, you need to determine how to improve your manipulation, then test new study participants.

Step 8: Report What You Did

•Researchers share what they did by presenting and/or publishing their research.

Characteristics of Science

•Science can be replicated. To replicate the findings means to discover a pattern of results similar to that obtained before. •Science is public. After conducting a research study, a scientist typically tells the scientific world about it by presenting the results at a scientific conference and/or publishing them in a scientific journal.

Characteristics of Science

•Science is empirical. Scientists rely on empirical data, evidence collected from the systematic observation or measurement of relevant information. •Science is objective, free from personal biases.

Summary

•Scientific method is a combination of empiricism and rationalism •Goals of science: description, prediction, explanation •Experimental method allows for explanation of cause-and-effect relationships •Testing a hypothesis involves attempting to falsify it

Characteristics of Science

•Scientists use the scientific method: a basic set of rules and procedures that govern the way research is to be conducted. •The general pattern: formulate a hypothesis, test the hypothesis, revise the hypothesis as needed, test again until data supports the hypothesis. Then you can form a conclusion.

PROBABILITY SAMPLING STRATEGIES (Survey Research)

•Simple random sampling -Each member of the population has an equal and independent chance of being chosen -The sample should be very representative of the population •Systematic sampling -Every Kth member of a population is chosen—e.g., every 10th name on a list. -More biased than simple random sampling. -Easier to select sample.

Small-N Designs

•Small-N designs (also referred to as single-participant designs or single-subject designs) are designed to help participants in changing behavior. They are called small-N designs because they use small samples; only one participant or just a few participants are tested. •In small-N designs, the behavior of a single person (or just a few people) is examined under well-controlled conditions to assess the effectiveness of a particular treatment on one or more target behaviors.

(Survey Research ch 7) Response Biases

•Social Desirability Effect—respondents may respond in a particular way because they fear their true response will not make a favorable impression on the observers. •Acquiescence (yea-saying)—this bias is present when people repeatedly choose "yes" or "strongly agree" regardless of item content. •Nay-saying- this bias is present when respondents tend to disagree with all statements. •Fence Sitting—respondents choose a response in the middle of the scale.

Hypothesis Testing: Statistical Significance and Errors

•Statistical significance (Alpha level) -An observed difference between two descriptive statistics (such as means) that is unlikely to have occurred by chance (sampling error) -We usually choose 5% level -This means 5% chance of making a Type I error •Which type of error is considered more serious by researchers? -Type I error -Cannot reduce both error rates at same time

Sources of Knowledge

•Superstition: knowledge that is based on subjective feelings, interpreting random events as nonrandom events, or believing in magical events •Intuition: knowledge gained without being consciously aware of its source •Authority: knowledge gained from those viewed as authority figures •Tenacity: knowledge gained from repeated ideas that are stubbornly clung to despite evidence to the contrary •Rationalism: knowledge gained through logical reasoning •Empiricism: knowledge gained through objective observations of organisms and events in the real world

Experimental Design with One Independent Variable: Analysis for One Independent Variable with Two Groups

•T-tests are commonly used to determine whether the difference between the mean scores of two groups is statistically significant. •If your two-group experiment uses a between-subjects design, then you will use a t-test for independent samples. •If your two-group experiment uses a within-subjects design, then you will use a t-test for dependent samples.

Tables and Figures

•Tables and figures are not required, but are often used. •For example, tables are typically used to provide an organized array of numbers such as means or correlations. •Figures, such as line graphs, are often used to illustrate statistical interactions. The APA Publication Manual also specifies how tables and figures should look.

Tables and Figures

•Tables and figures are not required, but are often used. •For example, tables are typically used to provide an organized array of numbers such as means or correlations. •Figures, such as line graphs, are often used to illustrate statistical interactions. •The APA Publication Manual also specifies how tables and figures should look.

Small-N Designs

•The ABAB design is another reversal design, and as before, "A" stands for baseline and "B" for treatment. •While the ABA design finishes with the participant back at baseline, the ABAB design finishes with the participant in the treatment phase. •If every time the treatment is taken away, behavior reverts back to baseline, then you can be confident the treatment was responsible for the behavior change. •Advantages of the ABAB design: •There are two opportunities to see behavior change as a result of the provided treatment (this is one reason why an ABAB design is preferred over an ABA design). We can think of an ABAB design as a replication of the experiment for a single subject; the second AB combination is a repeat of the first. •The ABAB design has the advantage of finishing the research study with the treatment still in place as opposed to leaving the participant at baseline. •There are some situations in which a reversal design should not be used: •1) if you are targeting a behavior that could not be reversed (such as riding a bicycle) •2) if you are targeting a behavior that should not be reversed (such as the taking of drugs)

A Few Words About Writing

•The APA Publication Manual is a good resource for recommendations regarding writing. Here are some pointers: •Use headings to organize your paper. •Use transitions to achieve continuity. •Use verb tense consistently within a given section. •Keep your writing free of bias. •Use the active voice when writing.

A Brief History of Research Ethics

•The Nuremburg Code—After World War II, physicians who had worked for the Nazis went on trial in Nuremberg, Germany for crimes committed against prisoners of war. •After the trial, judges created the Nuremberg Code (1947), a set of 10 ethical principles that must be fulfilled before medical experimentation can take place. •For example, the Nuremberg Code includes informed consent—decision to consent must be made with knowledge of what the experimental procedure entails. •The Tuskegee Syphilis Study—600 poor Black men were recruited in 1932 to be part of a research study run by the U.S. Public Health Service and the Tuskegee Institute. •Four hundred of the men had syphilis, a sexually transmitted disease. They were not told they had the disease, nor were they treated because the researchers wanted to learn about the short-and long-term effects of the disease. •The world beyond the medical community became aware of the Tuskegee Syphilis Study in 1972. •Knowledge of the Tuskegee Syphilis Study was a motivation behind The Belmont Report (1979) which covers ethical issues in biomedical and behavioral research. •The world beyond the medical community became aware of the Tuskegee Syphilis Study in 1972. •Knowledge of the Tuskegee Syphilis Study was a motivation behind The Belmont Report (1979) which covers ethical issues in biomedical and behavioral research. •The Belmont Report identified 3 primary ethical principles: •1) respect for persons—people are "autonomous agents," capable of making a decision about whether to participate in research. •2) beneficence—the research should deliver benefits to individuals and/or society. •3) justice—who should bear the burdens of research and who should benefit? Researchers need to be fair when considering these issues.

Abstract

•The abstract is on the second page of the manuscript. It is a brief summary of the manuscript and it should be written so that it stands alone (only information found in the manuscript can be included in the abstract). •The abstract is typically no more than 250 words and is typed in block format (no indentation). •The following information is generally found in an abstract: •Purpose of the research (the "purpose statement") •Number and type of participants •Brief description of the methodology •Brief description of the main results •A statement of the implications (what your results can mean for the real world) •Many journals also require that keywords be placed under the abstract. Keywords are words or phrases that represent main concepts from the manuscript. •Researchers can use keywords to find your article when searching the literature on PsycINFO.

Step 3: Choose a Research Approach

•The descriptive research approach is a collection of non-experimental methods (such as naturalistic observation, correlational research and surveys). •With this approach we are only observing, measuring and describing what we see. •Thus we cannot determine cause and effect with a descriptive research approach.

Discussion

•The discussion often starts with a restatement of the hypotheses and a statement of whether or not your hypotheses were supported. •You will also indicate whether you obtained any other findings that you consider to be important. •Then you want to discuss your results in light of previous research. Did you obtain results that are consistent with what others have found or consistent with what theory predicted? If not, why not? Provide possible reasons for the discrepancy. •The discussion often starts with a restatement of the hypotheses and a statement of whether or not your hypotheses were supported. •You will also indicate whether you obtained any other findings that you consider to be important. •Then you want to discuss your results in light of previous research. Did you obtain results that are consistent with what others have found or consistent with what theory predicted? If not, why not? Provide possible reasons for the discrepancy. •In a new paragraph, provide a critique of your research. Are there any methodological issues that should be addressed? Does your work have any limitations (for example, are you limited in your ability to generalize your results)? •You should provide ideas for future research. These ideas could come from your stated limitations and/or they could just be a reasonable next step in the research. •The last paragraph of the discussion often provides a statement of the implications of the findings and then provides a conclusion.

History of Small-N Designs

•The earliest researchers in psychology often studied single individuals, presenting the results for each rather than combining the data from multiple individuals. Additional individuals were tested to see whether the results would be replicated. •For example, Hermann Ebbinghaus studied his own memory, his ability to remember nonsense syllables of varying length over various delays. •Ivan Pavlov also worked with small samples, in fact, he worked with just one dog at a time. He and his colleagues found that a dog would salivate when seeing food, not just when the food was present in the dog's mouth. •So early on, small-N research was common. However, a change soon occurred with the development of inferential statistics, most notably the ANOVA. Researchers began to routinely test large groups of people, manipulating more than one independent variable, and testing for differences between the groups. •Small-N designs fell out of favor. •B.F. Skinner had been studying the behavior of animals such as rats and pigeons and determining how the consequences that followed behavior influenced subsequent behavior. He used very small samples in his work, and he became frustrated at how difficult it had become to publish his work. •So he and his colleagues created their own journal, primarily for the publication of experiments relevant to the behavior of individual organisms.

Step 3: Choose a Research Approach

•The experimental research approach is the most influential approach because it allows for the establishment of cause and effect. •In a simple experiment, researchers manipulate (vary) one variable (the independent variable) and observe the effects of that manipulation on a response measure (the dependent variable).

STRATIFIED SAMPLING (Survey Research)

•The goal of sampling is to select a sample that is representative of the population •But suppose— -That people in the population differ systematically along some characteristic. -And this characteristic relates to the factors being studied. •Then stratified sampling is one solution •The characteristic(s) of interest are identified (e.g., gender). •The individuals in the population are listed separately according to their classification (e.g., freshmen, sophomores, juniors, & seniors). •The proportional representation of each class is determined (e.g., 40% freshmen, 20% sophomores, 20% juniors, & 20% seniors). •A random sample is selected that reflects the proportions in the population, (e.g., 8 freshmen, 4 sophomores, 4 juniors, & 4 seniors.

THE FORMAT (Survey Research)

•The presentation is attractive, professional, and easy-to-understand •Questions and pages are clearly numbered •Directions are clear and explicit •Questions are objective •Questions are ordered sensibly •Transitions are used from one topic to the next

Chapter 6 Correlation Methods and Statistics Misinterpreting Correlations: The Third-Variable Problem

•The problem of a correlation between two variables being dependent on another (third) variable •More ice cream is eaten in summer, more drownings occur in summer, but those 2 to are not actually related to each other •Partial correlation -A technique that statistically removing the effect of the third variable from the correlation of the remaining two variables

Introduction

•The purpose of the Introduction section is to: •1) provide information about what others have done on the topic of interest •2) develop a rationale for doing the present research •3) provide general details regarding the present research. •A psychology research paper typically begins with a broad introduction to the general topic. •After a brief introduction to your general topic of interest, you need to summarize relevant research. •When presenting information regarding a source, note the author's/authors' last name(s) and the year the source was published. •When should you cite a source? Anytime you use someone else's ideas and/or words, you need to give them credit for the information. If not, you are plagiarizing. •In the last paragraph of the Introduction you should provide details about your specific research study. One way to do this is to include a purpose statement. Here is a suggested format for an experiment: •The purpose of the present experiment is to investigate the effects of [state first independent variable here with the levels in parentheses] on [state dependent variable summary here]. •Your purpose statement can be followed, in the same paragraph, by your hypothesis or hypotheses. Typically your hypotheses come from the information you obtained from previous literature. What does the previous research lead you to expect in your own work? •Then end this paragraph with a brief mention of your methodology.

Step 3: Choose a Research Approach

•The quasi-experimental approach resembles the experimental approach, but is missing one of the key components of experimentation: random assignment. •Thus we cannot determine cause and effect with a quasi-experimental approach.

THE QUESTIONS (Survey Research)

•The questionnaire contains questions that can be answered •The questionnaire contains questions that are straightforward

Step 9: Starting the Whole Process All Over Again

•The scientific process is cyclical! •Every study we conduct will likely answer some questions but also will likely lead us to ask additional questions.

Writing About and Presenting Your Research

•The sixth edition of the Publication Manual of the American Psychological Association (APA) governs the way we present our work. •Additional information is available from the APA style blog (blog.apastyle.org/) and tutorials are available at the APA website (www.apa.org). •A typical psychology research paper will contain the following sections: •Title Page •Abstract •Introduction •Method •Results •Discussion •References •Tables and figures (these are optional)

Step 3: Choose a Research Approach

•The small-N design approach involves testing just one or a few participants instead of testing large groups. •When you use a small-N design, you consider the results for each individual separately. •This design is often used in applied settings to determine whether a particular treatment works for a particular person. To achieve this, the researcher assesses the behavior of interest before (baseline) and after the treatment.

Small-N Designs

•The third type of small-N design is the changing criterion design. In this design a participant has to reach increasingly demanding goals over time. •Once baseline performance has been established, a target behavior (the criterion) is set and the participant is encouraged to meet that goal, with reinforcement to provide motivation. •Once behavior has reached the targeted level, the goal is advanced to a new level. •Thus, with this design a participant can become increasingly more adept at performing a target behavior (see next slide). Larwin and Larwin (2008) used a changing-criterion design to increase physical activity and to decrease media time for an adolescent. This figure illustrates the overall miles that the adolescent walked during baseline and two treatment phases.

Experimental Design with One Independent Variable: Manipulating Independent Variables

•There are a variety of way to manipulate an independent variable: •Presence/Absence of a Variable •Amount of a Variable •Type of a Variable •Presence/Absence of a Variable •For example: (see pic) •In some cases a placebo, an inactive substance, represents the active condition. •The condition in which the treatment is present is generally referred to as the experimental condition and the condition in which the treatment is absent is typically referred to as the control condition (note that a control condition doesn't have to be the absence of something).

RATING SCALES—TWO TYPES (Survey Research)

•Thurstone Scale Method of Equal Appearing Intervals •Statements indicating an attitude toward a topic are written •Judges rank the statements from least favorable to most favorable •Statements receiving consistent ratings are given the average score •A set of statements is selected that covers the entire range of attitudes •Here are some example items on peoples' attitudes toward the church (Scale values are indicated in parentheses). -I believe the church is the greatest institution in America today. (11) -I believe in religion, but I seldom go to church. (9.6) -I believe in sincerity and goodness without any church ceremonies. (6.7) -I believe the church is a hindrance to religion for it still depends upon magic, superstition, and myth. (5.4) -I think the church is a parasite on society. (.2) Source Thurstone, L.L., & Chave, E.J. (1929). The measurement of attitudes. Chicago: The University of Chicago Press. •Respondents check items with which they agree -Well-formed attitudes are indicated by consistently checking either high or low items -Poorly-formed or inconsistent attitudes are indicated by inconsistent pattern or by checking off many neutral items •Likert Scales •Statements indicating an attitude toward a topic are written •Items with clearly positive or negative attitudes are selected •Statements are listed with a space for respondent to indicate degree of agreement

Goals of Survey and Questionnaire Research (Survey Research)

•To evaluate people's beliefs/attitudes: e.g. -Beliefs about memory loss -Attitudes towards political candidates •To predict behavior: e.g. -Use of notes when shopping -Voting results

Small-N Designs

•Today, small-N designs are often used in applied settings to assess whether an experimental treatment affects one or more target behaviors. •There are different kinds of small-N designs: •1) reversal designs (also called withdrawal designs) •2) multiple-baseline designs •3) changing-criterion designs •No matter which design you use, you need to establish a baseline. Establishing a baseline means that you are assessing behavior prior to the presentation of the treatment. It's important to establish a baseline so you can determine if behavior has changed after treatment. •Since behavior can fluctuate, it's important to observe the behavior in question for an extended period prior to treatment in order to have a representative view of what the baseline behavior is like. Do not apply the treatment until you have established a stable baseline. •Researchers who use small-N designs typically use graphs to keep track of the behavior(s) in question. Graphing the behavior(s) will help you determine when the behavior you are using is stable (the line on the graph should be as horizontal as possible). •If your baseline isn't stable, try to identify the source of the variability and then try to control the influence. •Once your baseline is stable, it's time to provide the treatment. Again researchers generally use graphs to help them determine whether the treatment has influenced the behavior in question. •That's the general approach—now let's talk about specific small-N designs.

Hypothesis Testing: Type I and II Errors in Hypothesis Testing

•Type I error: an error in hypothesis testing in which the null hypothesis is rejected when it is true (groups only differed due to sampling error, but you conclude your treatment caused difference) •Type II error: an error in hypothesis testing in which there is a failure to reject the null hypothesis when it is false (groups actually differed, you said they didn't)

Experimental Design with One Independent Variable: Manipulating Independent Variables

•Type of a Variable •For example:

(Survey Research ch 7) Question Writing Guidelines

•Use easy-to-understand language. •Avoid writing questions that have negative wording. •Avoid writing double-barreled questions, which ask two questions in one. •Make sure your response options are mutually exclusive. •Avoid writing questions that contain emotionally charged language. •Consider providing your respondents with an "I don't know" response option.

POST-TEST ONLY CONTROL GROUP DESIGN

•Use when -Sample is sufficient (≥ 30/group) -Pre-testing is not possible •Disadvantages -If randomization is not effective, groups may not be equivalent -Cannot use pretest to assign to groups

Choosing A Sampling Method (Survey Research)

•What are the characteristics of the population of interest? •What sampling method will be most efficient and result in the most representative sample?

Identify a Sample (Survey Research)

•What are the characteristics of the population of interest? •What sampling method will be most efficient and result in the most representative sample?

SOLOMON FOUR GROUP DESIGN

•What if the pre-test interacts with the treatment? •The pre-test, post-test design will not work -You can't determine whether the treatment works by itself, or must include the pre-test •Many different comparisons are possible

UNSTRUCTURED QUESTIONS (Survey Research)

•What is your attitude toward government funding of the arts? •What do you think about American foreign policy?

QUESTIONNAIRES (Survey Research)

•What they are -Paper and pencil tests with structured questions -Self-administered •Advantages -Can be mailed out -Survey broad geographic area -Cheaper than one-on-one interview -Respondents may be more honest -Data easy to share with other researchers •Disadvantage Low return rate

Chapter 6 Correlation Methods and Statistics Misinterpreting Correlations: Curvilinear Relationships

•When a correlation coefficient does not adequately indicate the degree of relationship between the variables •The normal correlation techniques look for linear relationships. •Memory and age show an inverted U function, so r = 0, but they are still related, just not in a linear (straight line) function.

Chapter 5 Descriptive Research Methods Archival Research

•When a researcher conducts archival research, he or she is using the data from existing documents to answer research questions. These documents provide data collected by someone else. •Typically we analyze these documents using content analysis—a systematic analysis that results in a summary of the available information. •Advantages: •You do not have to expend the time, money or energy to collect the data yourself. The data are already available. •If necessary, you can supplement already-existing data with data you collect. •Disadvantages: •Obtaining access to archived data can sometimes be difficult. •No cause and effect.

(Survey Research ch 7) Nonprobability Sampling

•When random sampling is not possible, researchers often use nonprobability sampling—the researchers in this case will not know the probability of an individual's being selected for the study. •Because your sample is not random, it is not representative of the population and you need to be cautious about generalizing from the sample to the population of interest. •There are two types of nonprobability sampling: •1) convenience sampling •2) quota sampling •Convenience sampling (also known as haphazard sampling)—allows researchers to survey participants who are conveniently available (such as college students). •Quota sampling—starts with a convenience sample, but in this case, the researcher wants to ensure that certain types of participants are selected in particular proportions.

(Survey Research ch 7) Non-Response Bias

•When the population of interest is relatively large, surveying everyone in the population is impractical. So we typically just ask questions of a small portion of that group, a sample. Sampling is the process of selecting respondents for your study. •Two major sampling techniques: •1) Probability sampling •2) Nonprobability sampling

( ch 6) Correlational Research Correlation and Causality

•When two variables are correlated, we cannot conclude that one variable causes the other to occur because of: •1) the bidirectionality problem—we don't know whether X has caused Y or Y has caused X, and •2) the third-variable problem—a third variable may be causing both variables of interest. •Sometimes researchers try to remove the influence of possible third variables by measuring them when the data are collected and then calculating partial correlations, which statistically control for those variables in the analysis by "partialing out" or removing the third variable from the analysis. •Still can't conclude cause and effect.

( ch 6) Correlational Research Statistical Assessments of Relationships

•When you are assessing the relationship between two variables, you are calculating bivariate correlations ("bi" means two). There are several types. For example: •1) Pearson product-moment correlation (Pearson r), •2) Spearman correlation for ranked data, •3) point-biserial correlation •The type of data you are considering determines which type of statistic you use. In each case the correlation coefficient ranges from -1.00-+1.00 and is interpreted the same way. •A Pearson product-moment correlation (Pearson r) identifies the degree to which scores on two variables show a linear relationship. •The closer the Pearson r is to -1.00 or +1.00, the more linear the scores are, the more they approximate a straight line on a graph. •Scores used when calculating a Pearson r must be continuous variables (can take on any value between two numbers). •Another type of test for bivariate correlations is the Spearman correlation for ranked data. •This statistic can be used when the data are ordinal (ranked). •The point-biserial correlation is used when you want to calculate a correlation between one dichotomous variable (a variable with two choices such as Republican and Democrat) and one continuous variable. •A correlation coefficient can indicate whether there is a relationship between two variables. •Once you know that variables are significantly correlated, you may choose to do a regression analysis, which allows you to predict one variable when you know the other. •A regression analysis yields a regression equation which allows one to make the best possible prediction of scores. •The variable you are predicting is the criterion variable (also known as the y variable or the outcome variable). •The known variable is the predictor variable (the x variable). •When you use two variables in a regression analysis, the analysis is called simple regression. •A simple linear regression is a regression analysis that is linear—when you plot the data points on a scatterplot, they cluster around a straight line. •Multivariate statistics are data analysis procedures that are designed to analyze three or more variables at one time. •Multiple regression analysis is a statistical technique that allows us to look at the relationship between a number of predictor variables and a single criterion variable. •A multiple regression analysis tells us if the predictor variables are positively or negatively correlated with the criterion variable and it tells us the relative importance of each variable in determining the behavior of interest. •When you conduct a multiple regression analysis, you will get a multiple correlation coefficient (R) and it refers to the correlation between the combined predictors and the outcome variable. •If each of the predictor variables is correlated with the outcome variable, then your ability to predict the outcome variable will often be even greater when you use all the predictor variables together in the same analysis. •A multiple regression analysis will also give you a multiple coefficient of determination (R2) which refers to the amount of variation in the outcome variable that can be accounted for by all the predictors taken together. •The multiple regression analysis also allows us to determine the relative importance of each predictor variable to the outcome variable. The statistics are referred to as standardized regression coefficients or beta weights . •You can interpret beta weights (β), just the way you would interpret a correlation, and you can also test them for statistical significance just as you do with a correlation.

More on the Experimental Research Approach

•When you are conducting an experiment, the goal is to keep everything the same between your groups except for your independent variable. •We often use random assignment to place people into groups. When we do this, all characteristics of the participants are theoretically distributed across groups in approximately equal proportions (the groups are considered equal).

Consent for Minors

•When you are conducting research with minors, informed consent must be obtained from the participants' parents or legal guardians. •Minor who can read can also sign a form to consent to research. Agreement by a minor is called assent.

Experimental Design with One Independent Variable: Ceiling and Floor Effects

•When you are creating your dependent variables, you need to be concerned with ceiling and floor effects. •Ceiling effect—the independent variable appears to have no effect on the dependent variable only because the dependent variable task was too easy. To fix this, make the dependent variable task more difficult. •Floor effect—the independent variable appears to have no effect on the dependent variable only because the dependent variable task was too difficult. To fix this, make the dependent variable task easier.

More on the Experimental Research Approach

•When you can conclude that your independent variable caused the change in your dependent variable, your experiment is said to have high internal validity. •In other words, internal validity is the degree to which the results of an experiment can be attributed to the manipulation of the independent variable.

(Survey Research ch 7) Survey Research: The Terminology

•When you conduct a survey, you are attempting to estimate the opinions, characteristics, and/or behaviors of a particular population by asking a sample to respond to questions. •A population (sometimes called a target population or a population of interest) is a defined group of individuals. •A sample is a subset of individuals chosen from a population.

(Survey Research ch 7) Generalizing from the Sample to the Population

•When you do not randomly select your sample from the population, you are limited in your ability to generalize from the sample to the population. •Replicating research with different populations can help researchers to know how generalizable their results are.

Experimental Design with One Independent Variable: Analysis for One Independent Variable with More Than Two Groups

•When your experiment has more than two groups to compare, you should analyze your data using a one-way analysis of variance (ANOVA). This ANOVA can be a between-subjects ANOVA or a within-subjects ANOVA depending on your need. •A one-way ANOVA will allow you to determine whether there are significant differences between any of the group means. •If the ANOVA does not reveal that significant differences exist among your groups, then you would conclude that your independent variable did not affect your dependent variable. •If the ANOVA reveals that significant differences do exist among your groups, you need to use a post-hoc test to determine exactly which means are significantly different from each other.

Step 5: Recruit Study Participants

•Who to recruit? Your research question could help determine who your ideal participants should be. •One possibility is to use a convenience sample—typically, undergraduates. However, differences between your ideal population of interest and undergraduates can mean that results obtained with undergraduates may not generalize to your population of interest.

Reversal Designs

•With reversal or withdrawal designs, researchers establish a baseline, provide treatment and then remove it to determine whether behavior will revert to its naturally occurring state. •The simplest reversal design is the ABA design. "A" refers to the baseline condition—the target behavior in its naturally occurring state before the treatment. "B" refers to the treatment designed to alter the target behavior and the second "A" refers to the baseline again. •If the treatment was responsible for altering the target behavior, then taking it away should eliminate the behavior change. Here is an example of a graph of an ABA design. Note that multiple observations are used for both baseline and treatment phases. There is evidence of the treatment's effectiveness here—note the lack of data point overlap from one phase to the next and note that behavior went back to baseline once the treatment was removed.

Peer Review

•Work that is presented at a conference and published in journals tends to undergo peer review. This means that others who are familiar with the topic evaluate it and decide whether it is worthy of presentation/publication. •Peer reviews are often "blind." This means that those who are doing the reviewing do not know who did the research and those who did the research do not know who did the evaluation. •Rejection rates can be high. If your manuscript was rejected, you can send it somewhere else for possible acceptance (you may wish to make some of the evaluators' recommended changes before doing so).

Experimental Design with One Independent Variable: Increasing the Number of Levels of an Independent Variable

•You always need to have at least two levels of an independent variable, but you can have more than two. For example: •The number of independent variable levels that you decide to include in your experiment depends on the number of comparisons that you wish to make. Note that the more comparisons you make, the more participants you will need.

Method

•Your Method section is usually divided into a Participants subsection, a Materials or an Apparatus subsection and a Procedure subsection. Other options are a Design subsection and a Measures subsection. •In the Participants subsection, you should provide information about the participants such as the number tested, any information you collected about their characteristics (such as age), and any incentives you provided (such as extra class credit). •A Design subsection can be included in your manuscript when you have conducted an experiment. In this subsection you would provide information about your independent variable(s) and levels, and indicate whether your research was between-subjects, within-subjects or a mixed design. •A Design subsection can be included in your manuscript when you have conducted an experiment. In this subsection you would provide information about your independent variable(s) and levels, and indicate whether your research was between-subjects, within-subjects or a mixed design. •Researchers usually use either an Apparatus or a Materials subsection. •An Apparatus subsection is used to provide information about any type of equipment you use in your study (such as a computer to present stimuli). •A Materials subsection is often used to describe visual or verbal stimuli (such as vignettes). The details of a questionnaire can also be placed in this section. Some choose to use a Measures section to provide information about questions or scales they used with their study participants. •In a Procedure subsection you will describe all the major procedural steps in the order in which they occurred for your participants (including any procedures that you took to ensure your participants were treated ethically).

Method

•Your Method section is usually divided into a Participants subsection, a Materials or an Apparatus subsection and a Procedure subsection. Other options are a Design subsection and a Measures subsection. •In the Participants subsection, you should provide information about the participants such as the number tested, any information you collected about their characteristics (such as age), and any incentives you provided (such as extra class credit). •A Design subsection can be included in your manuscript when you have conducted an experiment. In this subsection you would provide information about your independent variable(s) and levels, and indicate whether your research was between-subjects, within-subjects or a mixed design. •Researchers usually use either an Apparatus or a Materials subsection. •An Apparatus subsection is used to provide information about any type of equipment you use in your study (such as a computer to present stimuli). •A Materials subsection is often used to describe visual or verbal stimuli (such as vignettes). The details of a questionnaire can also be placed in this section. Some choose to use a Measures section to provide information about questions or scales they used with their study participants. •Researchers usually use either an Apparatus or a Materials subsection. •An Apparatus subsection is used to provide information about any type of equipment you use in your study (such as a computer to present stimuli). •A Materials subsection is often used to describe visual or verbal stimuli (such as vignettes). The details of a questionnaire can also be placed in this section. Some choose to use a Measures section to provide information about questions or scales they used with their study participants. In a Procedure subsection you will describe all the major procedural steps in the order in which they occurred for your participants (including any procedures that you took to ensure your participants were treated ethically).

Lihat semua set pelajaran

Comprehensive final

Set pelajaran terkait

Finance 3000 Test 2

ESC1000 Midterm

Chapter 10- The Nervous System: Sensory Systems

Chapter 5 MasteringBio

CH. 65 Management of Patients with Oncologic or Degenerative Neurologic Disorders

Day 6

MUS 101 CH. 1- 12

NC Life Insurance Practice Exam Questions

The Historian's task in the Philippines

Exam 2- EAQ ?

ATI - Targeted Medical Surgical Neurosensory and Musculoskeletal

CPLP 9 & 10

Med Surg Hesi

Ch. 14 Marketing Set

PassAssured Test

HD412 Study Guide Exam 1

Integumentary system: C) Epidermal Derivatives

Nutrition Ch 1

World Geography Unit 3

Chapter 12 Cellular