PSY 355-Introduction to Experimental Psychology



PSY 355-Introduction to Experimental Psychology. J. P. Toth Fall 2007

Study Guide for Exam 2

Note: Students are responsible for all lectures and assigned material regardless of what is on this guide.

Lecture 6. Non-Experimental Methods [M&H, Chpt 3].

A. Overview of Non-Experimental Methods

1. Non-Experimental Methods: Why Use Them?

2. Quantitative vs. Qualitative data.

3. Internal vs. External Validity.

4. Degree of Manipulation & Degree of Constraint on Responding.

You should know be able to answer #1, define the terms in 2 & 3, and be able to locate non-experimental research, quasi-experimental designs, & true experiments on a (tricky?) graph of #4.

B. Phenomenology [cf. the Remember/Know Procedure].

C. Case Studies [cf. Patient CK].

D. Field Studies.

1. Reactivity & Unobtrusive Measures.

2. Naturalistic Observation.

a. Systematic Recording.

i. Frequency Method vs. Duration Method.

ii. Time Sampling vs. Event Sampling vs. Individual Sampling.

3. Participant-Observation Studies.

E. Archival Studies.

F. Qualitative Studies.

You should be able to define & identify examples of the methods above; and describe the main drawbacks to these methods. You should know how to define and identify examples of reactivity and unobtrusive measurements. You should know the difference between the various recording techniques.

Lecture 7. Surveys & Interviews [M&H, Chpt 4].

A. What are surveys and why are they used?

B. Constructing Surveys.

1. Closed Questions.

a. General Considerations.

i. Clarity & Simplicity.

ii. Leading & Double-Barreled Questions; Context Effects.

b. Semantic Differential.

c. Likert Scale.

2. Open Questions.

a. Content Analysis.

You should be able to answer A and define and recognize examples of each of the above terms (e.g., double-barreled questions). You should be able to identify or generate examples of the different question types (Open vs. Closed; Semantic Differential vs. Likert). You should know what content analysis is.

C. Measuring Responses.

1. Levels (or Scales) of Measurement.

2. Response Styles.

a. Willingness to Answer.

b. Position Preference.

c. Yea-sayers and Nay-sayers.

3. Manifest Content vs. Latent Content.

4. Reverse Scales and Reverse Scoring.

You should be able to identify the level of measurement requested by a question; and be able to give general definitions of the different response styles. You should know the difference manifest & latent content and know what reverse scales & scoring are, and why one would use them.

D. Sampling Strategies.

1. Probability vs Non-Probability Sampling.

2. Probability Sampling.

a. Simple Random Sampling.

b. Systematic Random Sampling.

c. Stratified Random Sampling.

d. Cluster Sampling.

2. Non-Probability Sampling.

a. Quota Sampling.

b. Convenience Sampling.

c. Purposive Sampling.

d. Snowball Sampling.

You should be able to define and identify examples of the different sampling methods. One way to study for this would be to examine, and generate alternative forms of, the examples given on the lecture slides.

D. Collecting Survey Data.

1. Mail Surveys.

2. Computer & Internet Surveys.

3. Telephone Surveys.

4. Self-Administered (Written) Surveys.

5. Interviews.

6. Focus Groups.

You should be able to define these different collection methods and know their main advantages & disadvantages.

Lecture 8. Correlations & Quasi-Experimental Designs [M&H, Chpt 5].

A. Correlation Basics.

1. Main uses of correlations.

2. General nature of the correlation coefficients (direction, strength, simple, scatter-plot).

3. Positive, Negative, & Zero Correlations.

4. A Variety of Correlation Coefficients.

5. Factors Influencing Correlations (Non-linearity, Restriction of Range, & Outliers).

6. Coefficients of Determination (r2) and Non-Determination (1-r2).

7. Why Correlation is not Causation (Directionality, Bi-Directionality, & Third Variables).

With one exception, you should know all of the above "chapter & verse" (i.e., very well). The one exception concerns the lecture slide on "A Variety of Correlation Coefficients"; for that slide, all you need to know is that different correlation coefficients (i.e., different calculation procedures) are used depending on the scale of measurement; you do not need to memorize the different types or what goes with what.

Note that I will likely ask you one or more questions that will require you to identify the direction of a correlation; these can be trickier than they first appear; if you are unsure, try creating a graph/scatterplot of the data with a best-fit-line before answering.

B. Prediction via Regression.

1. Bivariate Linear Regression.

a. Least Squares Regression Line (aka Line of Best Fit).

b. Y' = bX + a.

2. Advanced Topics.

a. Partial Correlation.

b. Multiple Regression (including R & R2).

c. Factor Analysis

i. as "data reduction".

ii. correlation matrix

iii. factor loadings (as compared to correlations).

d. Causal Modeling.

i. Cross-Lagged Panel Design.

ii. Path Analysis.

You should have a good working knowledge of Bivariate Linear Regression, including an understanding of the LSR line and the terms in the regression equation. For advanced topics, you should be able to provide a general definition of each technique (as provided on my slides and in the textbook). You should also (a) know the meaning of R & R2 in multiple correlation & regression; (b) be able to interpret a partial correlation;(c) know the difference between a Bivariate correlation & a factor loading; and (d) know what causal models are trying to achieve (duh).

C. Quasi-Experimental Designs.

1. Three Key Experimental Techniques that Allow one to Determine Cause & Effect.

2. Ex Post Facto Designs.

3. Longitudinal vs. Cross-Sectional Designs.

4. Pre-test/Post-test Designs.

a. One Group/Post-Test Only.

b. One Group/Pre-test/Post-Test.

c. Two Group/Post-Test Only.

i. Non-Equivalent Control Group Design.

d. Two Group/Pre-test/Post-Test.

You should know #1 and be able to define and identify examples of the different quasi-experimental designs; and be able to describe their main drawbacks (i.e., the things that make them 'quasi').

Lecture 9. The Basics of Experimentation [M&H, Chpt 7].

A. Main Components of an Experiment.

B. Overview of Experiments.

1. Independent Variables (IV).

a. Levels of the IV.

2. Dependent Variables (DV).

3. Experimental & Measured Operational Definitions.

C. Reliability.

1. Inter-rater Reliability.

2. Inter-item Reliability (aka. Internal Consistency).

a. Split-half technique.

b. Cronbach's alpha.

3. Test-Retest Reliability.

C. Validity.

1. Face & Content Validity.

2. Construct Validity

a. Convergent Validity.

b. Divergent Validity.

3. Criterion Validity

a. Concurrent Validity.

b. Predictive Validity.

D. Internal vs. External Validity.

1. Extraneous Variables & Confounds (see pp. 208-209 in the textbook).

E. Eight Classic Threats to Internal Validity.

1. History.

2. Maturation.

3. Testing.

4. Instrumentation.

5. Regression to the Mean.

6. Selection.

7. Attrition (aka Mortality).

8. Selection Interactions.

You should be able to define and identify examples of all of the above terms. Most important, however, is that you have an understanding of Reliability & Validity—what they mean and the different ways they can manifest themselves in data.

Discussion of Frank & Gilovitvh (1998)

You should know the general topic explored in this paper, the different methods employed to investigate that topic, and the general findings.

Material in the textbook & readings not covered in lectures

As far as I can tell, aside from the few things noted above, there are no major sections of material in the textbook that were not also covered in the lecture. Nevertheless, in addition to the lecture slides, I strongly encourage you to use the textbook and the textbook website to study for this exam.

Good Luck! I am really hoping that all of you do well.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download