PSYC300 - Exam 1 Review UMD

Active Deception

"commission"; the presenting of misinformation about the study to participants e.g. Study personnel tell subjects that they will be engaged in a cooperative task with other subjects, but instead subjects will actually be interacting with study personnel.

Constructs

- A type of variable in theories - Hypothetical entities created from theory - Cannot be observed or measured, but are assumed to exist - Help explain and predict behavior in a theory (examples: intelligence, hunger, heat) - Constructs must be linked to observable events that can be measured (e.g. hunger, productivity, anxiety)

Limitations of Operational Definitions

- An operational definition is not the same as the construct itself - Concerns about the quality of operational definitions and measurements produced - it is easy for operational definitions to leave out important components of a construct - Operational definitions often include extra components that are not part of the construct being measured. - Must be concerned with construct and measurement validity

How to improve internal validity

- Blinding: Participants—and sometimes researchers—who are unaware of what intervention they are receiving (such as by using a placebo in a medication study) to avoid this knowledge biasing their perceptions and behaviors and thus the outcome of the study - Random selection: Choosing your participants at random or in a manner in which they are representative of the population that you wish to study - Randomization: Randomly assigning participants to treatment and control groups, and ensures that there is not any systematic bias between groups - Study protocol: Following specific procedures for the administration of treatment so as not to introduce any effects of, for example, doing things differently with one group of people versus another group of people - Control for environment. Make sure that people are tested in the same lab at similar times, or all at the same time in a lecture hall, etc.

How to improve external validity

- Consider psychological realism: Make sure that participants are experiencing the events of a study as a real event by telling them a "cover story" about the aim of the study. Otherwise, in some cases, participants might behave differently than they would in real life if they know what to expect or know what the aim of the study is. - Replicate: Conduct the study again with different samples or in different settings to see if you get the same results. When many studies have been conducted, meta-analysis can also be used to determine if the effect of an independent variable is reliable (based on examining the findings of a large number of studies on one topic). - Try field experiments: Conduct a study outside the laboratory in a natural setting.

Strengths of the Correlational Research Strategy

- Describes relationships between variables - may be used when experimental method can't or shouldn't be used - may be easier to implement than an experiment - may be less intrusive -- natural behaviors - may have high external validity

Step 1: Empirical Research Question

- Observe behavior or other phenomena (leads to empirical research question) - Questions "that can be answered through the systematic observations and experiences that characterize scientific methodology" - Allows specific predictions

Empirical Questions Sources

- Personal interests and curiosities - Casual observation - Practical problems or questions - Behavioral theories - Reports of others' observations - What causes x? I think y causes x. - Academic publications

"Good" Scientific Theories

- Potential for falsification - must be able to test your theory - for a theory to be useful, the predictions drawn from it must be specific - specify what should happen - imply or specify what should not happen - Logical - Parsimonious - More precision (theory of knocking rhythms) - Quantity of confirmations - Quality of confirmations

Artifact

- Threats to both internal and external validity - external factor that may influence or distort measurements - experimenter bias - demand characteristics and participant reactivity

Theories

- an interrelated set of concepts (used to explain a body of data) - used to make predictions about the results of future experiments - an integrated set of hypotheses - specific predictions derived from theories based on empirical observation to be made - stated in terms of the operationalized variables of a particular experiment or study

Preventing Fraud

- experimental research studies that are designed to be replicated - research articles that are peer reviewed by experts in the field - consequences of being found guilty of fraud - reforms?

Correlational Research Assumptions

- normal population distributions - linear relationships between variables - at least ordinal level of measurement - homoscedasticity

Pseudoscience

1. A system of ideas often presented as science 2. Lacks some of the key components essential to scientific research e.g. astrology, phrenology, freud's psychoanalysis

Research Designs

1. Descriptive 2. Correlational 3. Observational 4. Experimental 5. Quasi-Experimental

Steps of the Scientific Method

1. Observe behavior or other phenomena - leads to empirical research question 2. Form a tentative answer or explanation (a theory) 3. Design an empirical test of your theory: how will you measure your variables? - operationism - level of measurement 4. Design an empirical test of your theory: what research design is best 5. Use your theory and research design to generate a hypothesis 6. Identify the participants or subjects for the study, decide how they will be selected, and plan for their ethical treatment 7. Conduct the study (get your data!) 8. Evaluate the data (statistics) 9. Interpret and report the results (Step 5 of NHST in APA format) 10. Refine or reformulate your research idea -- return to step 2 and continue Important: there are no final answers

Limits of Science

1. Requires repeatable and observable events 2. Many questions not able to be addressed - moral or ethical questions - May be used to support or assess morals - not able to answer which morals are empirically right

Writing out pearson's r and spearman's rho

1. restate the hypothesis 2. state what analysis was run 3. write the apa-style statistics (report to two decimal points) 4. include the means and standard deviations 5. evaluate the status of the hypothesis E.g. Mean self-esteem was 15.43 (SD = 2.11). Mean Depression was 10.96 (SD= 1.57). To test the hypothesis that self-esteem is correlated with depression, a Pearson's r was conducted. Self-esteem was significantly negatively correlated with depression, r(20) = -.74, p = .04. Therefore, the hypothesis that lower self-esteem would be correlated with greater depression was supported. ANOTHER e.g. We find a significant positive correlation between student motivation and exam scores (r(89) = 0.347, p < .05). P-value in the above example was listed as <.001 which is why there was no specific value and p <.05 was listed.

Writing out t-test

1.Restate the hypothesis 2.State what analysis was run 3.Write the APA-style statistics (reported to two decimal points 4.Include the means and standard deviations 5.Evaluate the status of the hypothesis E.g. To test the hypothesis that college students would be more stressed than high school students, an independent samples t-test was conducted. There was a significant difference between the two groups, t(75) = 2.11, p = .02, such that college students (M= 15.15, SD = 2.34) were more stressed than high school students (M = 11.75, SD = 2.53). Therefore, the hypothesis was supported. df = N - 1

Writing out Two-Way ANOVA

1.Restate the hypothesis 2.State what analysis was run 3.Write the APA-style statistics (reported to two decimal points 4.Include the means and standard deviations 5.Evaluate the status of the hypothesis E.g. To test the hypothesis that the relationship between intimate partner violence and PTSD symptoms depends on gender, a two-way ANOVA was conducted. There was a significant main effect of IPV on PTSD, F(1, 177) = 4.37, p = .03, such that those who experienced IPV had significantly greater PTSD symptoms (M = 9.33, SD = 3.02) than those who hadn't experienced IPV (M = 6.46, SD = 1.81). There was also a significant main effect of gender, F(2, 177) = 52.09, p = .02. A Tukey post-hoc test revealed that non-binary people had significantly greater PTSD symptoms (M = 10.70, SD = 2.83) than both men (M =6.00, SD = 1.83, p = .01) and women (M = 7.90, SD = 2.84, p = .01), while men and women did not significantly differ (p = .12). Finally, there was a significant interaction effect of IPV and gender on PTSD symptoms, F(2, 177) = 9.32, p = .01, such that non-binary people who experienced IPV had significantly greater PTSD symptoms (M = 13.20, SD = 1.30). Therefore, the hypothesis was supported.

The APA Ethics Code

10 ethical standards to provide guidance for psychologists with major focus on no harm, deception, informed consent, and confidentiality. Deception okay to use if during the debriefing, post-experimental explanation of the study to the participant, explains all information about the study and allows for informed consent.

Ceiling effect

A clustering of scores at the high end of a measurement scale. Allowing little or no possibility of increases in value.

Range effect

A measurement that is not sensitive enough to detect a difference

Systematic random sampling

A method of sampling in which sample elements are selected from a list or from sequential files, with every nth element being selected after the first element is selected randomly within the first interval e.g. As a hypothetical example of systematic sampling, assume that in a population of 10,000 people, a statistician selects every 100th person for sampling. The sampling intervals can also be systematic, such as choosing a new sample to draw from every 12 hours.

Theory

A set of statements about mechanisms underlying a particular phenomena

Correlation

A statistical technique that describes the relationship between two or more variables Variables are measured, with no manipulation by the researcher (ethical and pragmatic reasons)

Stratified sampling

A type of probability sampling in which the population is divided into groups with a common attribute and a random sample is chosen within each group. e.g. Let's say, 100 (Nh) students of a school having 1000 (N) students were asked questions about their favorite subject. It's a fact that the students of the 8th grade will have different subject preferences than the students of the 9th grade. For the survey to deliver precise results, the ideal manner is to divide each grade into various strata.

Method of Science

An approach to acquiring knowledge. Involves formulating specific questions and then systematically finding answers. Combines several different methods of acquiring knowledge

Empirical Science

Answers are obtained by making structured or systematic observations Results will provide support for the hypothesis or will not provide support.

Threats to external validity

Any characteristic of the study that limits the generality of the results Threats to generalizing across participants or subjects - selection bias - over-utilizing college students - volunteer bias - participant characteristics - cross-species generalizations threats to generalizing across features of a study - novelty effects - multiply treatment interference - experimenter characteristics threats to generalizing across features of the measures - sensitization -generality across response measures - time of measurement

Correlational Research Design

Assess linear relationship between two or more variables Data: measure two variables (two scores) for each individual in the group being studied Describes a relationship, but doesn't explain it Goal of correlation is determining association and strength of that association.

Bar chart

Bar charts are useful for displaying data that are classified into nominal or ordinal categories. Nominal data are categorized according to descriptive or qualitative information such as county of birth, or subject studied at university. Ordinal data are similar but the different categories can also be ranked, for example in a survey people may be asked to say whether they thought something was very poor, poor, fair, good or very good. Compares averages of both variables.

Weaknesses of the Correlational Research Strategy

Cannot assess causality (third variable problem, directionality problem, spurious relationships) Relief on self-report often (honesty of subjects, self knowledge of subjects, environmental factors) Low internal validity

Floor effect

Clustering of scores at the low end of a measurement scale. Allowing little or no possibility of decreases in value.

Interval and ratio scales

Consists of a series of equal intervals like the inches on a rule A ratio scale has a true zero value Generally continuous Compatible with basic arithmetic Suitable for most statistical analysis Interval example: Fahrenheit (there's no true zero, it can dip below 0). Ratio example: Test scores from 0 to 100.

Discrete variable

Consists of separate, indivisible categories. No values can exist between two neighboring categories.

Biserial

Continuous outcome Dichotomous predictor created from a continuous variable E.g. Outcome: measuring romantic relationship satisfaction with a survey asking on a 0 - 100 continuous scale how satisfied people are with their romantic relationships Predictor: A 100 item communication survey provides a communication score between 0 - 500. People who score 250 or less are classified as low communication and people who score 250 or greater are classified as high communication Biserial correlation is almost the same as point biserial correlation, but one of the variables is dichotomous ordinal data and has an underlying continuity. For example, depression level can be measured on a continuous scale, but can be classified dichotomously as high/low.

Point Biserial

Continuous outcome Truly dichotomous predictor E.g. Outcome: measuring romantic relationship satisfaction with a survey asking on a 0 - 100 continuous scale how satisfied people are with their romantic relationships Predictor: first or last sibling* no only children are included ANOTHER E.G. Researchers want to assess whether there is a relationship between type of exercise (running vs. yoga) and mood measured on a 0-100 scale with equal intervals.

Artifact of Demand Characteristic

Demand characteristics refer to any of the potential cues or features of a study that suggest to the participants the purpose and hypothesis of the study, and/or influence the participants to respond or behave in a certain way. Causes participant reactivity. Participants modify their natural behavior in response to knowing they are in a study.

Major components of the Results section

Demographic Information, Re-Stating the Hypothesis, Statistical Analysis, and Statement of Support/Non-Support Objective voice should be used in Results/Discussion section.

Nonexperimental research strategy

Demonstrates a relationship between variables

Validity

Did you measure what you said you measured? Measurement procedure must accurately capture the variable that it is supposed to measure Construct validity (scores obtained from a measurement behave exactly the same as the variable itself) Face validity (whether a measure superficially appears to measure what it claims to measure) Predictive validity (scores obtained from a measure accurately predict behavior according to a theory) Concurrent validity (scores obtained from a new measure directly related to scores obtained from a more established measure of the same variable) Convergent validity (strong relationship between the scores obtained from two or more different methods of measuring the same construct) Divergent validity (showing little or no relationship between the measurements of two different constructs)

Observational research design

Direct observation to assess Linear relationship between two or more variables (data: observe two variables for each individual in the group being studied) Group differences (data: observe two a score for each individual in two or more groups)

Institutional Review Board (IRB)

Each institution or agency is required to establish and IRB committee - composed of scientists and nonscientists - examines proposed research involving humans - reviews research proposals according to seven criteria 1. minimization of risk to participants 2. reasonable risk in relation to benefits 3. equitable selection 4. informed consent 5. documentation of informed consent 6. data monitoring 7. privacy and confidentiality

Science

Evidence is gathered from careful, systematic, and public observations

Threats to Internal Validity

Extraneous variables - any variables in a research study other than the specific variables being studied Confounding variables (confounds) - extraneous variables (usually unmonitored) - change systematically along with the independent variable being manipulated - provide an alternative explanation for the observed relationship between the two variables Environmental variables: general threats for all designs Assignment bias: participant-related threats for designs that compare different groups. Between subjects design. Time-related variable: threats for designs that compare one group over time. Within subjects design.

Reactivity: roles participants may take

Good subject role (supports the experimenter's hypothesis) Negativistic subject role (acts contrary to the hypothesis) Apprehensive subject role (presents himself in a "good light") Faithful subject role (follows instructions to the letter - ideal participant)

Historical Highlights and the Belmont Report (1979)

Identifies three basic ethical principles for protecting human subjects 1. Respect for persons (individuals are autonomous and some require protection) 2. Beneficence (minimize harm and maximize benefits) 3. Justice (equitable division of benefits)

Cluster sampling

In cluster sampling, researchers divide a population into smaller groups known as clusters. They then randomly select among these clusters to form a sample. Cluster sampling is a method of probability sampling that is often used to study large populations, particularly those that are widely geographically dispersed. Researchers usually use pre-existing units such as schools or cities as their clusters. e.g. Consider a scenario where an organization is looking to survey the performance of smartphones across Germany. They can divide the entire country's population into cities (clusters) and select further towns with the highest population and also filter those using mobile devices.

Reliability

Is your measure consistent in what it measures? Stability or the consistency of the measurements produced by a specific measurement procedure. Inconsistency of a measurement comes from error. Measured Score = True Score + Error Successive measurements: test-retest reliability compares score of two successive measurements of the same individuals and correlates the scores Simultaneous measurements: Inter-rater reliability - agreement between two observers who simultaneously record measurements of the behaviors Internal consistency: split-half reliability - splits the test in half computing a separate score for each half, and then calculating the degree of consistency between the two scores for a group of participants. Reliability is a prerequisite for validity (a measurement procedure cannot be valid unless it is reliable). It is not necessary for a measurement to be valid for it to be reliable. Can be reliable without validity. You can have reliability without validity, but cannot have validity without reliability.

Statistical Abbreviations

M = mean SD = standard deviation N = total number in sample n = number in subsample p = statistical significance r = pearson's correlation rs = spearman's rho t = t-test f = ANOVA

Experimental research strategy

Manipulation of a variable and random assignment Produce a cause-and-effect explanation for the relationship between two or more variables Data: create two treatment conditions by changing the level of one variable; then measure a second variable for the participants in each condition YOU CAN'T MANIPULATE RACE so can't do that for an experiment.

Limiting the artifact of demand characteristics and reactivity

Naturalistic observation Deception Between subjects designs

Nuremberg "Doctors" trails

Nazi party doctors who did experimentation on prisoners in concentration camps

IRB Submission categories

No IRB needed - no interaction with subjects - no identifiable information collected - no publication intentions Exempt - interaction, but minimal risk Expedited - minimal risk with some other qualifications such as no anonymity (drug trials, blood samples, behavioral interferon's) Full review - greater than minimal risk

Quasi-experimental research strategy

No random assignment. Can never produce an unambiguous explanation. Attempt to produce a cause-and-effect explanation, but not clear causation. Data: measure before/after scores for one group that receives a treatment and for a different group that does not receive the treatment.

Public Science

Observations are available for evaluation by others Replication is key to the scientific method

Nonprobability sampling

Odds of selecting a particular individual are unknown

Spearman's rho or Kendall's tau

Ordinal level data for outcome or predictor or both If one variable is not ordinal, it must be continuous E.g. Outcome: self-report survey asking how satisfied people are in their romantic relationships (1 = not at all happy, 3 = blissfully happy) Predictor: number of words spoken to their romantic partner each day) or Predictor: 1 = no communication, 3 = we never stop communicating ANOTHER E.G. Researchers want to assess whether there is a relationship between time of day (morning, afternoon, and evening) and hunger measured on a 0-100 scale with equal intervals.

Pearson's r

Outcome and predictor are continuous E.g. Outcome: measuring romantic relationship satisfaction with a survey asking on a 0 - 100 continuous scale how satisfied people are with their romantic relationships Predictor: measuring communication with a survey asking on a 0 - 100 continuous scale how much people communicate with their romantic partner ANOTHER E.G. Researchers want to assess whether there is a relationship between time spent studying in hours and exam score (0-100).

Phi

Outcome and predictor are dichotomous E.g. Outcome: people report if they are happy or not in their romantic relationships Predictor: people report if they communicate or not in their romantic relationships ANOTHER E.G. Researchers want to assess whether there is a relationship between political party (Democrat or Republican) and voting behavior (voted or didn't vote in the last election.

Logistic

Outcome is dichotomous Predictor is discrete (usually > 2) E.g. Outcome: People report if they are happy or not in their romantic relationships Predictor: People report if they have low, medium, or high levels of communication in their romantic relationship

Requirements of the Discussion section

Overview, Interpretation, Limitations, and Future Directions/Implications

Sources of Error

Participant changes (the participant can change between measurements) Environmental changes (it is difficult to attain the ideal of identical circumstances) Observer error (the individual who makes the measurements can introduce human error)

Participant bias

Participants will sometimes second-guess what the researcher is after, or change their answers or behaviors in different ways, depending on the experiment or environment.

Tense for scientific writing (excluding proposal)

Past tense

Descriptive Research Design

Produce a description of individual variables as they exist within a specific group Obtains a snapshot (a description) of specific characteristics of a specific group of individuals Data is usually in the form of averages or percentages

References page

References should be double spaced with a handing indent, alphabetical order, and the references heading should be centered. E.g. Eisenberg, D., Gollust, S. E., Golberstein, E., & Hefner, J. L. (2007). Prevalence and correlates of depression, anxiety, and suicidality among university students. American Journal of Orthopsychiatry, 77(4), 534-542. https://doi.org/10.1037/0002-9432.77.4.534

Conclusions of NHST

Reject the null hypothesis if your p-value is less than your alpha value (.05). Supports your alternative hypothesis. Fail to reject the null hypothesis if your p-value is greater than your alpha value (.05). Fails to support your alternative hypothesis.

Ordinal Scale

Represents differences in a series of ranks Discrete Ordered descriptions. E.g "i am unhappy." "i am ok" "i am awesome" E.g. 1. If I give each person a rank based on when they enter a classroom 2. If you arrange your professors based on who appears tallest to shortest 3. If I measure pre-date anxiety by having people rate if they are "Doing okay", "Mildly concerned", "Freaking out", and "Panic!!"

Nominal Scale

Represents qualitative differences in the variable measured Discrete Unordered descriptions. E.g "i'm a turtle" "I'm a snail" "i'm a butterfly" E.g. 1. If a local animal shelter keeps track of the breed of dogs at the shelter 2. If I record whether customers at a local McDonald's order a diet or regular drink with their food

Scatterplot

Scatter plots' primary uses are to observe and show relationships between two numeric variables. The dots in a scatter plot not only report the values of individual data points, but also patterns when the data are taken as a whole. Identification of correlational relationships are common with scatter plots. e.g. researchers survey 40 participants in the midst of an officially declared pandemic, like COVID-19, on their altruism using a previously validated altruism scale that ranges from 0 - 50 and is assumed to have equal intervals. They then ask how many times participants how likely they are to wear a mask when they next go out during the pandemic on a 0 (will definitely not wear a mask) to 100 (will definitely wear a mask), assuming equal intervals.

Running head

Should be a short title (50 characters or less) starting in the left corner and the page number in the right corner

Methods Heading

Should be bolded and centered Participants subheading should be bolded, but not underlined or italicized and should be kept on the left. Required subheadings: participants, measures, procedure, and design

Selecting your Correlation Coefficient

Smoking and lung cancer - Pearson's r: if lung cancer is measured by number of lung cancer cells present. if smoking is measured by number of cigarettes smoked per day. Spearman's rho: if lung cancer is measured in stages (stage 1, stage 2, stage 3, etc.). If smoking is measured by number of cigarettes smoker per day (or in ordinal scales such as never smokes, moderate smoker, heavy smoker, etc.) Point (Biserial): if lung cancer is measured by number of lung cancer cells present. If smoking is measured by being a smoker or not. Logistic: If lung cancer is measured as being present or not. If smoking is measured on a scale of never smokes, occasionally smokes, moderate smoker, heavy smoker Phi: if lung cancer is measured as being present or not. If smoking is measured as a smoker or non-smoker

Hypotheses

Specific predictions of research Stated in operational terms Based on your theory (may be one or two directional) Consistent with design (experimental or non-experimental) Null hypothesis example: there is no relationship between time spent solving a common task and romantic relationship satisfaction. H0: ρ≥ 0 Alternative hypothesis example: there is a negative linear relationship (correlation) between time spent solving common tasks and romantic relationship satisfaction. H1: ρ< 0 Both of the above are based in probability and the alternative hypothesis is primarily stated in research.

NHST Review

State your hypotheses (null and alternative) State your criterion (alpha - type 1 error, one or two-tailed test, critical region) Collect your data Compute your statistics Evaluate and interpret your results

stem and leaf plot

Stem and leaf plots are useful in some cases because you can see where the bulk of scores lie. In the above graph, most scores were in the 20s or 60s. Bar charts also show this information, but the advantage the stem and leaf plot have is that you can see all of the scores (other charts usually show just totals).

Informed consent

Subject must be informed of all available information about the study so they can make a rational decision to participate (or not). Consent form contains all the elements of informed consent and a place for the participant to sign.

Probability sampling

The exact size of the population must be known and it must be possible to list all the individuals. Each individual in the population must have a specified probability of selection The selection process must be unbiased; must be a random process

Fraud in science

The explicit effort of a researcher to falsify or misrepresent data Why do researchers commit fraud? - to remain competitive in their academic environment - "to help science" - "publish or perish"

External validity

The extent to which the results of a research study can be generalized - generalization from a sample to the general population - generalization from one research study to another - generalization from a research study to a real-world situation e.g. a goal of their study was to "illuminate whatever differences may currently exist between the content of male and female graffiti, at least within the populations from which our sample is drawn."

Histogram

The histogram is used for variables whose values are numerical and measured on an interval scale. It is generally used when dealing with large data sets (greater than 100 observations). A histogram can also help detect any unusual observations (outliers) or any gaps in the data.

Proportionate Stratified Sampling

The population is subdivided into strata. Number of participants from each stratum is selected randomly. The proportions in the sample correspond to the proportions in the population. Guarantees the sample will be perfectly representative of the population. May be unfair to minority strata. May not fulfill rule of all participants having an equal chance. e.g. will produce a sample that will be representative of the population in at least one pre-defined characteristic (such as race, gender, religion, etc.) (the same as proportional stratified random sampling)

Titple page

Title in center of page, not bold, authors full name one to two spaces below the title. Then affiliation, course name and number, instructor name/title, assignment due date

Requirements for Tables and Figures

Title, Labels, Error Bars, Description, and Legend (if necessary)

Directionality problem

Two variables, X and Y, can be statistically related because X causes Y or because Y causes X.

National Research Act

U.S. Surgeon General (late 1960s) - orders federally funded research proposals from the Public Health Service to be reviewed for ethical treatments of subjects National Research Act (1974) - mandated regulations for protection of subjects

Confidentiality

Usually ensured through anonymity Includes: - attitudes and opinions - measures of performance - demographic characteristics

Nuremberg Code (1947)

Voluntary informed consent Foal should be positive results for society Previous data justification for research Avoid unnecessary physical and mental suffering No implied risk of death or injury Risks in proportion to benefits Protections against participant risks Researchers should be trained and scientifically qualified Participants able to quit at any time Medical researchers must stop if continuation would be dangerous

Appendix

Where you should provide a screenshot/text of all survey questions. Appendix goes after the references. Each appendix should be mentioned within the text, and each should start on a new page. Example of citing the first appendix: (Endicott et al., 1993; Appendix A)

In-text citation

Year of the article should come immediately after the author(s) name

Experimenter bias

a phenomenon that occurs when a researcher's expectations or preferences about the outcome of a study influence the results obtained

Continuous variable

a quantitative variable that has an infinite number of possible values that are not countable

Spurious correlation

an apparent but false relationship between two (or more) variables that is caused by some other variable

Internal validity

concerned with factors in the research study that raise doubts or questions about the interpretation of the results. a research study with high internal validity produces a single, unambiguous explanation for the relationship between variables internal validity and external validity are a trade-off. Need balance of the two.

Operationism

defining a concept in terms of the specific procedures used to represent it we assume operational scores from participants reflect true score and error. e.g. A researcher is interested in better understanding the relationship between pet owners and depression as previous research has indicated that pet ownership may increase feelings of security and life satisfaction. The researchers survey 50 families who report if they own pets or not and complete a depression inventory that is scaled on a 1 - 100 scale and assumed to have equal intervals. Outcome construct: depressions Operation of the construct: depression survey

Passive Deception

omission of information, keeping secrets e.g. The study involves covert procedures, such as subjects being observed behind a one-way mirror. The subjects' ignorance regarding their observation is intended to affect their behavior.

Proportional stratified random sampling

probability based sampling method where the study population is divided into subgroups based on key characteristics, and the subjects are selected from subgroups so the fraction of the sample consisting of each subgroup is the same as the fraction of the subgroup in the study population e.g. A research team has decided to perform a study to analyze the grade point averages or GPAs for the 21 million college students in the U.S. The researchers decide to obtain a random sample of 4,000 college students within the population of 21 million. The team wants to review the various majors and subsequent GPAs for the students or sample participants. Out of the 4,000 participants, the breakdown of majors is as follows: English: 560 Science: 1,135 Computer science: 800 Engineering: 1,090 Math: 415 The researchers have their five strata from the stratified random sampling process. Next, the researchers study the data of the population to determine the percentage of the 21 million students that major in the subjects from their sample.

Convenience quota sampling

subgroups are identified to be included. quotas are established for individuals to be selected through convenience form each subgroup. allows a researcher to control the composition of a convenience sample. the sample probably is biased.