Short answer Question 1 Information

Short answers Questions

1 .For class, we read a paper by Good and colleagues (2012), which focused on validating a measure of belonging in academic domains. Across all the studies, the research found that their new sense of belonging scales and the five subscales (membership, acceptance, affect, trust, and desire to fade) had good Cronbach's alphas. Moreover, in Study 2, the authors had participants complete the new sense of belonging scale at two time points and found that the scale at time 1 correlated strongly with the scale at time 2 (r= .87). In Study 2, the authors also found that the sense of belonging scale significantly predicted important outcomes including - math anxiety, usefulness of math, math confidence, and intent to pursue math. Moreover, the sense of belonging scale predicted these outcomes over and above other outcomes, including psychological sense of school membership, general anxiety, and math identification. Please answer the questions below based on these findings, the paper, and general course concepts.

https://psycnet.apa.org/record/2012-02597-001

a. What does it mean if a scale has good reliability? How did the researchers assess reliability in this paper?

b. What does it mean if a scale has good construct validity? Name two ways research can assess construct validity (that we discussed in class).

c. How did Good and colleagues examine the construct validity of their scale?

d. Table 6 in the paper reports the findings from a multiple linear regression model and reportsthe standardized regressioncoefficients. Remember, this model included the sense of belonging (the main predictor of interest, psychological sense of school membership, general anxiety, and math identification). Looking at intent to pursue math, sense of belonging had a standardized regression coefficient of 0.42,p< .001. Provide an interpretation of this regression coefficient. What does this number mean?

2. For class, we read a paper by Chaney and Sanchez (2021), which focused on validating a measure of prejudice confrontation styles. Across all the studies, the researchers found that their new prejudice confrontation styles and the five subscales (membership, acceptance, affect, trust, and desire to fade) had good Cronbach's alphas. Moreover, in Study 2, the authors had participants complete the prejudice confrontation styles at two time points. The researchers found that all the subscales at time 1 correlated strongly with their time 2 scores. In Study 2, the authors also found that the prejudice confrontation styles measure predicted important outcomes, including - life satisfaction and autonomy. Moreover, the prejudice confrontation styles measure predicted these outcomes over and above other outcomes, including participant gender, race, discrimination experience, confrontation frequency, intercultural conflict styles, and coping with discrimination scale. Please answer the questions below based on these findings, the paper, and general course concepts.

https://journals.sagepub.com/doi/full/10.1177/13684302211005841

a. What does it mean if a scale has good reliability? How did the researchers assess reliability in this paper?

b. What does it mean if a scale has good construct validity? Name two ways research can assess construct validity (that we discussed in class).

c. How did Chaney and Sanchez examine the construct validity of their scale?

d. Table 5 in the paper reports the findings from a multiple linear regression model and reportsthe unstandardized regressioncoefficients. Remember, this model included the prejudice confrontation styles subscales (the main predictors of interest) and participant gender, race, discrimination experience, confrontation frequency, intercultural conflict styles, and coping with discrimination scale. Looking at autonomy as the outcome, the prejudice confrontation styles help-seeking subscale had unstandardized regression coefficient of 0.16,p= .01. Provide an interpretation of this regression coefficient. What does this number mean?

3. Dr. Veronica Johnson is a researcher interested in helping effects. She wonders whether people will help more on sunny days than cloudy days. She also wonders whether people will help women more or less often than men. Over a period of two months, She and her research team go out on 10 different cloudy days and 10 different sunny days. At an outdoor mall, one member of her research team approaches individuals and asks for a ride to a gas station because their car is out of gas. On half the trials, a woman research team member approaches the participant, and on the other half of the trials, a man research team member approaches the participant. The team finds that on sunny days, the man is helped 75% of the time and the woman is helped 60% of the time. On cloudy days, the man is helped 15% of the time, and the woman is helped 60% of the time. Please answer the questions below based on these findings and general course concepts.

a. What are the independent and dependent variables for this study? (Note there is more than one IV.)

b. Briefly describe the study design. Be sure to describe whether the independent variables are within or between-subjects.

c. Does there appear to be a main effect of sunny vs. cloudy days on helping behavior? What is the mean difference between helping on sunny vs. cloudy days (i.e., the mean of helping on sunny days minus the mean of helping on cloud days)?

d. Does there appear to be a main effect of confederate gender on helping behavior? What is the mean difference between the woman vs. man confederate (i.e., the mean of helping a woman minus the mean of helping a man)?

e. What is the simple effect of helping a woman on sunny versus cloudy days? That is, what is the mean difference between helping on sunny vs. cloudy days among women (i.e., the mean of helping on sunny days minus the mean of helping on cloud days for women)?

f. What is the simple effect of helping a man on sunny versus cloudy days? That is, what is the mean difference between helping on sunny vs. cloudy days among men (i.e., the mean of helping on sunny days minus the mean of helping on cloud days for men)?

g. Based on your answers to 1e and 1f, does it look like there is a significant interaction?

4. For class, we read a paper by Moss-Racusin and colleagues (2012), in which they randomly assigned women and men STEM faculty members to evaluate a male or female candidate for a lab manager position. Everything was the same about the application; the only thing that differed was the name - John vs. Jennifer. For this question, we will focus on the results for student's worthiness of mentorship (these results are presented in Table 1). The researchers found that men faculty rated John an average of 4.74 for mentorship and rated Jennifer an average of 4.00. Moreover, the researchers found that women faculty rated John an average of 4.73 for mentorship and rated Jennifer an average of 3.91. Please answer the questions below based on these findings, the paper, and general course concepts.

https://www.pnas.org/doi/10.1073/pnas.1211286109

a. What are the independent and dependent variables described above? (Note, you do not need to list all the DVs listed in the actual paper.)

b.Briefly describe the study design. Be sure to describe whether the independent variables are within or between-subjects.

c. Based on the study design, what is the most appropriate statistical test for this study?

d. Does there appear to be a main effect of John vs. Jennifer on mentorship? What is the mean difference between John vs. Jennifer on mentorship (i.e., the mean for John minus the mean for Jennifer)?

e. Does there appear to be a main effect of faculty gender on mentorship ratings? What is the mean difference between the woman vs. man STEM faculty (I.e., the mean for woman faculty minus the mean for men faculty)?

f. What is the simple effect of mentorship ratings for John vs. Jennifer among women faculty? That is, what is the mean difference between John vs. Jennifer days among women STEM faculty? (I.e., the mean for John minus the mean for Jennifer among women STEM faculty?)

g. What is the simple effect of mentorship ratings for John vs. Jennifer among men faculty? That is, what is the mean difference between John vs. Jennifer days among men STEM faculty (I.e., the mean for John minus the mean for Jennifer among men STEM faculty)?

h. Based on your answers to 1e and 1f, does it look there is a significant interaction?

5. Imagine Dr. Pietri wants to try a new belonging intervention for psychology majors at CU Boulder. To initially test this intervention, she decided to try it in introduction to psychology courses. Specifically, Dr. Pietri delivered the belonging intervention in intervention class and compared it to another class, which served as the control (control class). Dr. Pietri administered a measure of belonging at the start and end of the semester among students in the control and intervention course. Thus, Dr. Pietri could look at changes in belonging in both classes. At time 1 (before any intervention), Dr. Pietri finds that students in the intervention class have a belonging score of 2.70, and students in the control class have a belonging score of 2.80. At time 2, Dr. Pietri finds that students in the intervention class have a belonging score of 4.00, and students in the control class have a belonging score of 3.00. Please answer the questions below based on these findings and general course concepts.

a. Briefly describe the study design. Be sure to describe whether the independent variables are within or between-subjects.

b. Based on the study design, what is the most appropriate statistical test for this study?

c. At time 1, does there appear to be a significant difference between the intervention class and the control class? What is the mean difference between the intervention class and the control class at time 1 (I.e., the mean for the intervention class minus the mean for the control class at time 1)?

d. At time 2, does there appear to be a significant difference between the intervention class and the control class? What is the mean difference between the intervention class and the control class at time 2 (I.e., the mean for the intervention class minus the mean for the control class)?

e. Based on your answers to d and c, does it look like the intervention was successful? Why or why not?

f. Based on the methodology used by Dr. Pietri, can she conclude anything about causation? Put another way, was this a true experiment? Why or why not?