Research Methods Final Exam

Methods of Sampling 1. ________ sampling - the odds of selecting a person are known and can be calculated >>>Have to know size of population and be able to list all people >>>Each person must have a specific probability of being chosen >>> Everyone assigned same probability 2. _______ sampling - odds of selecting a specific person unknown 3. ______ _____ sampling - everyone has an equal chance of being selected Sampling with replacement vs. sampling without replacement 4 _______ sampling - similar to random sampling, but start at a random place and then choose every nth person 5 ______ _______ sampling - identify subgroups and then select equal-sized random samples All same proportion, you have a certain number of males and females in your society, but all the same amount 6 ______ ______ random sampling - use stratified random sampling but make sample sizes proportionate to population Based on percentage, instead of making it all equal. You would base it on populations percentage 7. ______ sampling - selecting large groups together (still random but as group rather than individual) 8. ________ sampling - use participants who are easy to get 9. ______ sampling - stratified sampling but not random

1 probability 2 nonprobability 3 simple random 4 systematic 5 stratified random 6 proportionate stratified random 7. cluster 8. convenience 9. quota

Research Strategies: 1. _______ Research Strategy: not concerned with relationships between variables 2. _______ compare individual scores, do not look at differences between groups as with other research strategies 3. _______ examines cause and effect relationships between variables 4. _______ use per-existing groups to measure outcomes 5. _______ tries to establish relationships between variables but doesn't attempt to explain relationship

1. Descriptive (Ex: On average, students at the local college spend 12.5 hours studying outside of class each week and get 7.2 hours of sleep each night.) 2. Correlation (EX: amount of time spent studying per week and GPA) >>>cannot determine a cause and effect relationship 3. Experimental (EX: Increasing the amount of exercise causes a decrease in cholesterol levels.) 4. Quasi - experimental (EX:The treatment may cause a reduction in smoking behavior, but the reduced smoking may be caused by something else.) >>>often but not always use in cases in which doing a true experimenter would be unethical or impossible 5. Non-experimental (ex: There is a relationship between gender and verbal ability. Girls tend to have higher verbal skills than boys, but we don't know why.)

Internal Validity Issues (Within Subject Designs) 1. Confounding from ___________ 2. Confounding from ___________ 3. _____ - environmental variables other than the treatment that change over time 4. _____ - change in participants (often developmental but not necessarily) that occurs during treatment and may affect study results 5. _____ - changes in a measuring instrument 6. ____ _____ ___ - tendency for extreme scores to move closer to the mean 7. _____ ____ - when the experience of participating in one treatment influences the scores in later treatment conditions 8. ____-____ _____ - when one treatment produces a change in scores that carry over to other treatment groups 9. _____ ____ - changes in participant behavior related to participating in a study but not related to treatment, such as practice effect and fatigue

1. environmental (Textbook gives example of time of day, which is a bit confusing as it is different from the concept listed below) 2. time related variables (Refers to amount of time participated in study/order of variables) 3. History 4. Maturation 5. Instrumentation 6. Regression towards the mean 7. Order effects 8. Carry-over effects 9. Progressive error

Between Subjects Design Vs. Within Subjects Design 1. Within Subjects Design requires ______ participants 2. You can _____ ______ for individual differences in a within subjects design 3. _____ ______ is an issue with Within Subjects Designs- participants drop out 4. Choosing Within Vs. Between _____ _____- can be eliminated with Within Subjects Design _____ _____- Between Subjects Design can eliminate this _____ ______- need slightly less with Within Subjects Design 5. Can also use a _____ ______ ______, which attempts to use aspects of both- separate treatment conditions, but attempt to approximate the same participant in each group

1. fewer 2. statistically control 3. participant attrition 4. - individual differences - time-related factors - fewer participants 5. matched subjects design

What's a Factorial Design? 1. Factor = _____ _______ 2. Levels = ______________________________ (basically how many groups are you creating for the one variable) 3. Two-factor design = _____________________

1. independent variable 2. number of values you are creating for the independent variable 3. two independent variables

Applied vs Basic ______ research - research that looks at theroetical issues; done for the sake of doing research _______ research - research to look at practical issues or solving real-life problems

Basic Applied

______ ______ - involves direct observation of behavior and systematically recording behaviors you want to study (e.g. observing children on the playground for frequency of sharing)

Behavioral Observation

_____ ______ a combined strategy uses some experimental vs. nonexperimental strategies (An example would be also looking at education level (college vs. no college) as a possible factor in memory performance in the previously mentioned example (this would then make the design a 2x2x2 design))

Combined strategy (different from mixed strategy)

______ ______ is a name used to describe the dependent variable in a variety of statistical modeling contexts, including multiple regression

Criterion variable

_____-_____ - compare two different groups of people at different ages

Cross-sectional

Bias in Research ______ _________ - potential cues that suggesst to the participant what the hypothesis is and therefore influences behavior >>> ________ effect - participants alter behavior becasue they know they are being observed (reactivitiy) >>> ______ _______ - people present themselves as more positive than they are >>> _______ ______ - be as close to the same with everybody as you can

Demand Characteristics >>> Hawthorn effect >>> Social Desirability >>> Experimenter Expectation

______ ______ are brief descriptive coefficients that summarize a given data set, which can be either a representation of the entire or a sample of a population.

Descriptive statistics

Possible Problems with Between Groups Designs _____ _______ - people may drop out of the study based on the groups they are in (E.g. comparing healthy controls to a patient population; the patient population is much more likely to drop out) _____ ______ - spread of treatment from the experimental group to the control group _____ ______ - if a control group learns about a treatment, they may also demand to take part in the treatment ____ _____ - control group works extra hard to look like the experimental group ____ ____ - people give up if they learn another group is receiving special treatment

Differential attrition Diffusion Compensatory equalization Compensatory rivalry Resentful demoralization

Methods of quantifying observation: _____ _____ - counting number of times a behavior occurred within a set time period ______ ______ - how much time does someone spend engaging in a particular behavior ______ _____ - divide observational period into a series of time intervals and recording if the behavior occurs in that interval or not

Frequency method Duration method Interval method

______ -______ _____ ______ - The basic concepts of a two- factor research design can be extended to more complex designs involving three or more factors; >>>> A three- factor design, for example, might look at academic performance scores for two different teaching methods ( factor A), for boys versus girls ( factor B), and for first- grade versus second- grade classes ( factor C).

Higher- Order Factorial Designs

_____ ______ use a random sample of data taken from a population to describe and make inferences about the population.

Inferential statistics

APA Ethics Code: 6 Principles ____ consent : this is the most important part of research! This is what you use to inform potential participants about your study. Participants should generally be well informed about what your research involves (though you do not need to disclose your specific hypothesis) understand the information provided to them, and also choose to participate

Informed (look at photo)

The primary purpose of the _______ is to protect the rights and welfare of human subjects involved in research activities being conducted under its authority.

Institutional Review Board (IRB)

_______ - one factor has a direct influence on a second factor, causing differences in the means that are not explained by the main effects (An interaction would be if the performance in the dim-light and background noise condition was much worse than could be explained by either variable alone)

Interaction

_______ scales often used in survey research- use numerical ratings on a scale with descriptors, can use quantitative statistical techniques

Likert

________ - measure same group of participants over time

Longitudinal

______ ____ - Effects from just the independent variables manipulated (Referring to the previous example, a main effect would be that background noise in both lighting conditions caused worse performance on the memory task)

Main effect

____ ____ - some combination of between-subjects and within-subjects design (in example, one group of participants gets the bright vs. dim condition with noise, the other group of participants gets bright vs. dim conditions without noise)

Mixed Design

Belmont Report; 3 main underlying ethical guidelines 1. Principle of ______ - this shows respect for individuals as a person to consent to participate in research, as well as withdrawal consent) 2. Principle of _____ - do no harm. Should minimize risks and maximize benefits from study 3. Principe of _____ - costs and benefits should be distributed equally to all participants in a fair manner

National Research Act 1. autonomy 2. beneficence 3. justice

Nuremberg Code - Developed based on _______ experiments 10 Important points that all research must follow 1. _____ consent is essential as well as the _____ capacity to provide _______. ____ choice is all essential 2. _______ shoudl be to improve society or be beneficial to society in some way 3. _______ should be designed in a way so that previous research on the subject, both based on human and animal research demonstrate a _______ of the condition being studied and the potential results ______ the experiment. 4. the experiment should avoid unnecessary ______ and ______ suffering 5. No studies shoudl be conducted where ___ or _____ _____ is likely to occur 6. Level of _____ should never exceed the level of ______ that would be gained from the study 7. proper precautions should be taken to prevent against any remote possibilities of ___ or ____ 8. Experiments should only be conducted by those ______________ ___________ 9. __________ should be allowed to leave or end the experiment 10. Scientist should be prepared to _____ the experiment at any point if determined that continuation would cause greater harm than the potential benefits of completing the study

Nazi 1. Voluntary; legal; consent; free 2. research 3. research; knowledge; justify 4. physical and mental 5. death or serious injury 6. risk; knowledge 7. death or injury 8. scientifically qualified 9. participants 10. end

Scales of Measurement ______ scale - qualitative differences in the variables measured (gender, major, etc) _____ scale - rank ordered variables (who completes a test first to last without considering difference in time) ____ scale - organized sequentially and all categories are the same size, no true zero ____ scale - same as interval scale except there is a true zero

Nominal Ordinal Interval Ratio

_______ ______ - measures the degree and the direction of the linear relationship between two variables

Pearson correlation

Pearson correlation coefficient

Pearson's r +1 and −1, where 1 is total positive linear correlation, 0 is no linear correlation, and −1 is total negative linear correlation

Relationships between Variables ______ relationship - when one variable increases, the others increase ______ relationship - when one variable increases, the other decreases

Positive Negative *there can also be no relationship between the variables*

How to Limit Confounding Variables (Between Subjects Designs): - ____ ______ - some random process is used to assign participants to groups, - _____ _____ _____ - want to try and keep groups as the same number of participants as much as possible so limit group assignment (once total number is reached in one group, everyone left must go in the other) - _______ - assigning individuals to groups so that a specific participant variable is balanced across the group to make the groups as equivalent as possible - _______ ____ ______ - can also restrict study to only a certain number of factors, such as using only female participants if gender is likely to be a confounding variable

Random assignment Restricted Random Assignment Matching Holding variables constant

________ is a statistical measurement used in finance, investing, and other disciplines that attempts to determine the strength of the relationship between one dependent variable (usually denoted by Y) and a series of other changing variables (known as independent variables).

Regression

________ is defined as the consistency of measurement > Does not measure accuracy

Reliability

What's the difference between an experiment and a quasi-experiment?

The experimenter is actively changing the independent variable, unlike in the quasi-experiment. Quasi-experiment has no random assignment

Placebo

any inert substance or technique that causes no biological effect (began in medical research but has also moved into psychology including "placebo" psychotherapy)

A ___________ is a actor involved in a psychological experiment who appears to be a fellow participant to other research participants but who is actually working for the researcher in some way.

confederate

Controlling Confounding Variables (Between Subjects Designs): ______ _____ - should use same process to obtain participants as is possible ______ _____- all participants should receive exactly the same treatment except for variables manipulated or controlled for by the researcher Composed of ______ ______ - characteristics of one group should be as similar to those in the other group as is possible

created equally treated equally equivalent individuals

_____ _____ - variable observed for changes based on changes made to the independent variable by the experiment

dependent variable

Counterbalancing(Within Subjects Designs) - Want to have _____ numbers in each possible order of treatment conditions - Does not eliminate _____ _____ Complete counterbalancing vs. partial: - Easy to counterbalance only two groups but gets more difficult as possible combinations _______, with partial counterbalancing, ensure each treatment condition is ______

equal order effects increase; first

_______ validity- how much can results of this study be generalized to the population _____ validity- there is one and only one explanation for the results of the study

external internal

External Validity: How well can the research be generalized from the sample included in the study to the _______ _______ How well can the research be generalized to _____ _____ how well does the research apply to ________-______ ________

general population (A sample of all female college freshman probably has low external validity for the general population) other studies (if the same study conditions are present can it be replicated) real world problems

_____ ______ - the variable manipulated by the experimenter in a study

independent variable

The _____ is the average of a data set. The _____ is the most common number in a data set. The ______ is the middle of the set of numbers.

mean mode median

Types of Validity To establish validity, you must demonstrate that the measurement procedure is actually ______________________________________________. ____ validity : apparent on surface what the test is measuring _____ validity : showing that scores are highly correlated with an already established measure of the construct ____ validity : how well does the construct predict actual behavior ___ validity : scores demonstrate that measurement of the construct and the construct itself behave the same ____ validity : involved two different ways of measuring the same construct that are highly correlated ____ validity : shows two constructs are distinct because there is no correlation

measuring what it claims to be measuring Face validity (how happy you are on a scale from 1 - 10) Concurrent vallidity (how well does something you are measuring compare to something already established) Predictive validity (a memory test and how well it is actually measuring a certain behavior) Construct validity (how well are the ways you are measuring getting at the construct) Convergent validity (when you want to establish two things as the same thing IE: 2 different IQ tests) Divergent validity (when you want to show 2 things are not measuring the same thing IE Memory vs IQ test ) ***you cannot have validity without reliability; have to have reliability first, before you can have validity***

Internal Validity - study has high internal validity if there is ______________________; if someone could easily state another reason your study results could have happened, the study has low internal validity

only one possible explanation of the study results

a _______ is any numerical quantity that characterizes a given population or some aspect of it. This means the parameter tells us something about the whole population.

parameter

Deception in Research _____ deception : withholding information about the study (omission) ____ deception : presenting misinformation about the study to participants (commission) If deception is used, ______ is required

passive acitve debriefing

Population vs Sample _____ - everyone within a certain group, can refer to all humans, all people of a gender, all people who have a certain job, etc. _____ - a subset of the population you wish to study _____ _____ - whatever group the researcher is interested in _____ _____- people who are able to be recruited for a research study

population sample target pop. accessible pop.

Linear relationships can be ____ or ____

positive or negative

A ____________ is a variable used in regression to predict another variable. It is sometimes referred to as an independent variable if it is manipulated rather than just measured.

predictor variable

Primary vs Secondary Sources ______ Source - written by the person who actually conducted the research _____ Source - description of another person's study

primary secondary

Different Effects ____ effect - when scores cluster towards one end of the scale and you are not able to measure properly _____ effect - scores are all too high (everyone getting 100%) ____ effect - scores are all too low

range ceiling floor

When we ___________, the two sample values are independent. Practically, this means that what we get on the first one doesn't affect what we get on the second. Mathematically, this means that the co-variance between the two is zero. In _________________, the two sample values aren't independent. Practically, this means that what we got on the for the first one affects what we can get for the second one.

sample with replacement sampling without replacement *For more info* https://web.ma.utexas.edu/users/parker/sampling/repl.htm

Between Subjects Designs Only one ______ per participant (this doesn't mean you measure only one variable, it means the person can't participate more than once in the same group or be assigned to both groups) Need to have a ______ sample size as each group should have a decent number (n of ~30) _______ ______ can lead to variance ______________________ can also lead to variance

score large Individual differences Environmental factors that change systematically

Threats to External Validity __________ - not an accurate representation of population (a lot of psychological research is based on college students and is probably not generalizable to the population)

selection bias

You can attempt to control bias through blinding: _____-blind : participant does not know group _____ - blind : neither participant nor experimenter know group

single double

_____ _____ looks at how spread out a group of numbers is from the mean, by looking at the square root of the variance.

standard deviation

Types of Reliability _____-_____ reliability - administer same test to same person on several occasions ____ - _____ reliability - use different versions of same test _____-_____ reliability - degree of agreement between two separate raters ___-____ reliability - measure of internal consistency

test-retest *.700 is good parallel-forms *two different measures that look at the same thing inter-rater * two different people are making observations on something/someone and will compare scores afterwards *.75 is the rule of thumb split-half *split measure in half and compare scores *.7 is ok, .8 is great, and anything higher than .95 is redundant

Methods to Control Issues in Within Subjects Designs Controlling/limiting the amount of _____ of the study can limit some confounding factors, such as maturation Change study design to a _______ subjects design _______ - different participants undergo treatments in different orders

time; between; counterbalancing

The ______ measures the average degree to which each point differs from the mean—the average of all data points

variance

Two basic experimental designs: ________ design - different sets of data all obtained from same participants (only one set of participants) _________ design - different data sets from different groups of participants

within - subjects between - subjects