Discussion questions for Rosenthal and Rosnow, Essentials of Behavioral Research: Methods and Data Analysis, 3rd edition.

        Chap 1  Chapter 1 Spirit of Behavioral Research (p. 3)

  1. Relate "why" and "how" questions to experimental and descriptive research and give an example in either counseling or I&O psych to illustrate the two.
  2.  Describe how the concept of motivation can be different or the same in the contexts of counseling and I&O psychology settings.
  3. Which of Charles Peirce's "paths to understanding" would religion rely upon?
  4.  Describe a relational study with humans that could be done to support Harlow's findings for the importance of affect on healthy development of children?
  5. Why are empirical rules or principles characterized as "probabilistic assertion" when used by behavioral scientists? 

    Chap 2 Contexts of Discovery and Justification (p. 37)
  1. Does the method of empirical inquiry enable scientists to determine if there is a god? Why or why not?
  2. What is the relationship between theory and hypothesis?
  3. What was the importance of Brady's serendipitous finding?
  4.  Give an example of an operational definition for the theoretical notion of "need for approval."
  5. How would you change a p value to reduce the likelihood of a Type II error?
  6. How is effect size influenced by the size of the study and the size of the statistic?
  7. Why is it important to know the effect size of a finding?

        Chapter 3 Ethical Considerations, Dilemmas, and Guidelines (p. 61)

  1. Can you "puzzle" a way to justify using children as opposed to monkeys to develop new anti-viral drubs?
  2. Could Gardner's research (1978) have been conducted not using deception?
  3.  Is surreptitious observation of elevator riders an unethical use of surveillance?
  4. Is it ethical to conduct research on elderly demented patients if consent is given by their power of attorney? 
  5. Give an example of conducting research when the cost is higher for doing it than not doing it, and another in which the cost is higher for NOT doing it.   

        Chap 10 (REVIEW) Describing and Displaying Data (page 293)

  1. What is a sampling unit? Give an example of two other than an individual.
  2. Define variance and standard deviation
  3. Why does the formula for sample variance use N-1 in the denominator?
  4. What is the purpose of squaring a deviation score?

        Chap 4 Reliability and Validity of Measurements (page 87) 

  1. How is the effective reliability of judges influenced by the number of judges?
  2. How many judges would be needed to achieve an effective reliability of .85 if the mean reliability was r = .45?
  3. Why is  % agreement among judges misleading and what would you use as an alternative? 
  4. What type of reliability is replication considered to be?
  5. Give an example of convergent and discriminant validity for constructs in your project.
  6. What's the difference between internal and external validity? 

        Chap 11 (REVIEW) Correlation (page 314):

  1. What is the purpose of the BESD? Give an example of how it could be used with your project.
  2. Give an example of how both a point-biserial and a phi coefficient could be used with your project? 
  3. What are two statistics that can be used to test the significance of phi?
  4. Give an example of a predicted curvilinear relationship between two variables from your project.

        Chap 6 Interviews, Questionnaires, &  Diaries (page 160)

  1. Cite two comparative advantages and two disadvantages for using an (1) interview and (2) questionnaire.
  2. What is an adverse outcome that could result from not carefully following each of the four steps to developing an interview schedule?
  3. How does the randomized response technique guarantee confidentiality and reduce socially desirable responding?
  4. Is there any reason to believe that assuring confidentiality could actually increase non-response rates?
  5. Is yea-saying a problem with the F-scale or G-scale of the MMPI?
  6. What it the problem with using propositional items in cross-cultural research and how would you avoid it?
  7. What do you think would be the most important trait to include in a semantic differential measure used to evaluation a teacher?
  8. How is inter-judge reliability determined when the Q-sort is used to assess the personality of someone?
  9. In developing a true Likert scale, why would items with low item to total score correlations not be used?
  10. Would a uni-dimensional or multi-dimensional scale be better for measuring an attitude? Why?
  11. What is a major advantage of using a self-recorded diary over retrospective memory to obtain autobiographical data?
  12. What is the problem with using "Do you approve or disapprove of the way the way Mayor O'Malley is handling his duties?" as an interview question?

        Chap 13 (REVIEW) Comparing Means (page 381)

  1. How is the magnitude of the significance test affected by the effect size and size of the study? Why?
  2. Explain how each of the three ways of maximizing t cause it to increase.
  3. What is the difference between the t for independent and non-independent samples?
  4. Explain each of three assumptions for errors when using the t test.
  5. How is t related to F and what are the advantages of each when comparing means?
  6. Why is eta used as an effect size estimate for F?
  7. Why is F a ratio and why does it have two sets of degrees of freedom?
  8. How does the Bonferonni procedure protect the researcher from doing something wrong?

            Chap 7  Randomized Controlled Experiments and Causal Inference (page 189)

  1. What's another name for sampling unit? Give an example of a sample unit other than an individual.
  2. What's the difference between a between-subjects design and a repeated-measures design?
  3. What kind of randomization is essential for a true experimental design and why?
  4. Give an example of each of the five classes of IV relating to your project. (biological, environmental, hereditary, training and experience, maturity.
  5. Give an example, using your project, to illustrate the "three imperatives" (Hume's classical rules).
  6. Why does concomitant variation not ensure causation?
  7. What are the greatest threats to internal validity in your study?
  8. What is a potential demand characteristic you would need to control for in your project?
  9. How do the following strategies: quasi-control subjects, preinquiry, blind controls, and sacrifice groups serve to control for demand characteristics?
  10. Which of the strategies (see Table 6.2, p. 123) would be most useful for controlling subject-artifact influences that may jeopardize internal validity in your experiment?
  11. Should Kinnebrook have been fired because he was "too slow"? 
  12. What is the difference between an experimenter-expectancy effect and a demand characteristic?

        Chapter 9 Randomly and Nonrandomly Selected Sampling Units (p. 260)

  1. What is bias and instability in surveys? 
  2. Are more sampling units required for a heterogeneous or homogenous population?
  3. Is it more important to have a representative sample or random sample?
  4. Why does stratification increase the likelihood that a sample will be representative?
  5. Can increasing sample size reduce non-response bias?
  6. What is a volunteer characteristic that could bias a sample chosen for testing a hypothesis in your study?
  7. How can you reduce the magnitude of non-response bias?
  8. What argument could you make to justify using a fortuitous sample?

       Chapter 14 Analysis of Variance and the F test (p. 409)
 

        Chap 16 Factorial Design of Experiments (page  343)

  1. What are the two primary reasons for conducting a factorial design?
  2. How many factors and levels of each factor are there in a 4X5 factorial design?
  3.  What are interaction, or residual effects that are left over after the row and column effects have been partitioned from the grand mean?
  4. Is the source of error in a factorial design derived from the individual differences within each condition or within each treatment (factor)?
  5. Why is eta used to estimate the effect size in factorial designs?
  6. Describe a situation where you may want  to test the grand mean for some hypothesis relating to your project.
  7. Given that the effect size is not zero, will the t statistic increase or decrease as each the samples sizes across conditions becomes more equal? (tale 16.5)
  8. In the ANOVA table 16.8 what effects are statistically significant?

        Chap 17 Interaction Effects (page 365)

  1. Which finding regarding the sex and teaching method study is accurately portrayed in figure 17.1 and which is not? Why?
  2. What conclusion regarding the results (fig 17.1) could be inferred that may be misleading?
  3. What is the appropriate conclusion depicted accurately in figure 17.2?
  4. How many interactions are possible in a A X B X C design? What are they?
  5. How may two-way interactions are there in a four-way design and how many three-way interactions?
  6. How are  row effects and column effects defined?
  7. How does Table 17.6 differ from Table 17.5?
  8. In the more complex 3X4 (therapy by types of treatment) design, which type of patient improved least under which treatment? Which types of patients benefited most?
  9. In the results of the three way design, sex X drug X psychotherapy, would males be better off without the drug?
  10. Are the benefits of the drug less for males than for females?
  11. Which sex benefited more from psychotherapy?
  12. What is a synergistic effect and how does it differ from interaction?
  13. What is the difference between  positive and negative synergistic effects?
  14. What is a quadratic interaction. A cubic interaction?  

          Chapter 18 Repeated-Measures Designs  (page 392)

  1. Does "nested" or "crossed" describe a repeated measures design?
  2. Why is repeated measures referred to as a "within" subjects design?
  3. What is the advantage of using subjects as their own controls?
  4. In the design illustrated in Table 18.1, which source of variance is of interest to the researcher, the column or row effects?
  5. What is the difference between how between subjects variance is used in a randomized (group) design and a repeated measures design?
  6. What is the difference between fixed and random effects?
  7. What is the purpose of a Latin Square design?
  8. Give an example of a study using a repeated measures design that would require counterbalancing.
  9. What is the difference between sequence and order?
  10. Give an example of a 2 (between) X 2 (within) design using variables in your project.
  11. What could an intraclass r be used for in your project? 

        Multiple Regression

      Chapter 21 Meta-Analysis: Comparing and Considering Research Results (p. 663)