Exercise 2: reliability and validity
200 words needed.. I have to do my part
For this exercise, your task is to estimate the reliability and validity of a measure of Need for Cognition (nCog; Cacioppo & Petty, 1982; Cacioppo, Petty, & Kao, 1984). This is a team project. You have been randomly assigned into groups of 3-4 classmates. Please work with your team members and turn in one final, written document with all classmates’ names on it by the assigned due date.
An SPSS data file is included in the assignment folder, with responses from 294 college students. Within the data file you will find raw scores for nCog items, as well as composite scores for several other variables (see page 2 for descriptions).
Your tasks for this assignment are described below. To answer each of these questions, run the analyses using SPSS and report relevant information in your write-up (i.e., the reliability or validity coefficient). As a team, write a report explaining your findings for each of the questions above. Provide an interpretation of what the results mean. In other words, explain your findings in sufficient detail so that another person could understand what you examined and why.
Estimate Coefficient Alpha for nCog using the items in the data set.
Estimate test-retest reliability for nCog – data were collected 2 weeks after the original nCog data (NCOG_T2).
Report the criterion-related validity estimate for nCog using College GPA (CollegeGPA).
Report the criterion-related validity estimate for nCog using Student Satisfaction (StudentSat).
Report the incremental validity estimate of nCog above and beyond High School GPA (HS_GPA) and ACT Scores (ACT) in predicting College GPA (CollegeGPA).
Report the incremental validity estimate of nCog above and beyond High School GPA (HS_GPA) and ACT Scores (ACT) in predicting Student Satisfaction (StudentSat).
Report convergent validity evidence using Intrinsic Motivation (IntMot) and Mastery Goal Orientation (MGO).
Report discriminant validity evidence using Extrinsic Motivation (ExtMot), Performance Approach Goal Orientation (AppGO), Performance Avoid Goal Orientation (AvdGO), and Social Desirability (SocDes).
As a reminder, this is a team project. You will be randomly assigned to teams and you must work with your classmates to complete this assignment and turn in one single document for everyone. You can coordinate your efforts using Google Hangouts, or any other method you prefer. Google Docs works great for writing the paper together as a team. You already have access to both of these using your UMSL email/SSOID. You can download SPSS for free as an UMSL student, or you can use TritonApps to run your analyses remotely. Please note that a part of this assignment is evaluating your team members’ contributions. If you do not contribute (as indicated by your team members), your grade will be adjusted accordingly. You cannot work on this assignment individually.
Variables in Data Set/Assignment
Need for Cognition –described as a drive to understand situations and the world around us (Cacioppo & Petty, 1982; Cacioppo et al., 1984). People with a high need for cognition tend to enjoy problem solving, abstract reasoning, and enjoy complex situations/problems. Items were rated on a 5-point scale (1 = Strongly Disagree to 5 = Strongly Agree), and higher scores indicate higher levels of the construct. Items were taken from Cacioppo et al. (1984), in the attached article. Note that several of these items are reverse-scored. You will need to re-code them prior to running the analyses.
High School GPA – is cumulative, high school grade point average, on a 4.0 scale. Scores were taken from admissions records.
ACT Scores – ACT scores were gathered from university records.
College GPA – is cumulative, college grade point average on a 4.0 scale. This was taken from university records. In this study College GPA was collected one year after Need for Cognition was measured.
Student Satisfaction – is the extent that students are satisfied with their experiences at the university. Six items were written for the purpose of conducting this study. Cronbach’s alpha for this scale was .90. In this study student satisfaction was collected one year after Need for Cognition was measured.
Academic Motivation focuses on the reasons that students go to college. This is based on Deci and Ryan’s (1985) self-determination theory (Ryan & Deci, 2000). For the purposes of this data collection ten items were written to measure intrinsic motivation and ten items were written to measure extrinsic motivation.
Intrinsic Motivation – is based on a drive to learn and accumulate knowledge. The process of learning itself is internally motivating for individuals with a high level of intrinsic motivation. Coefficient alpha for this scale was .92.
Extrinsic Motivation – is based on the anticipation of some other, external reward, Individuals with a high level of intrinsic motivation are less interested in learning itself, but motivated because with a college degree they are more likely to obtain a high-paying job or promotion at work. Coefficient alpha for this scale was .87.
Goal Orientation – reflects the purpose, or type of goal that a person aims to achieve when pursuing a particular task (Elliot et al., 1999). This can be composed of mastery goal orientation, performance approach goal orientation, and performance avoid goal orientation. Seven items were written to measure each of the three dimensions below.
Mastery Goal Orientation – has a focus on improving oneself as a function of task pursuit, and is often associated with deep-level processing and drive to learn and understand material (e.g., expanding one’s knowledge). Coefficient alpha for this scale was .90.
Performance Approach Goal Orientation – is associated with achieving positive outcomes and successful task accomplishment. However, this is more often associated with surface-level processing, with a focus on simply obtaining task outcomes (e.g., getting a good grade). Coefficient alpha for this scale was .83.
Performance Avoid Goal Orientation – is also focused on achieving outcomes, but specifically on avoiding failure or negative outcomes. Instead of focusing on learning material or getting a favorable grade or expanding one’s knowledge, the focus remains on not failing a class, for example. Coefficient alpha for this scale was .89.
Social Desirability – involves providing what would be perceived as favorable, or socially desirable, responses (Crowne & Marlowe, 1960). The KR20 internal consistency estimate for this scale was .87.
Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and Social Psychology, 42, 116-131.
Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of need for cognition. Journal of Personality Assessment, 3, 306-307.
Crowne, D.P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349–354.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York, NY: Plenum.
Elliott, A. J., McGregor, H. A., & Gable, S. (1999). Achievement goals, study strategies, and exam performance: A mediational analysis. Journal of Educational Psychology, 91, 549-563.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78.
Me and my group will divide reliability, criterion-related validity, construct-related validity. one person has choosen to complete criterion-related validity. So left is reliability and construct-related validity. you can choose which everone you will like.
This is what one of my partners have completed so far.
Hi everyone, here’s a draft of my section of the project. I’m working on explaining the results a bit better, and am going back through the previous studies to see how these results compare/diverge from the others. Feedback would be appreciated! -Kaitlin
Criterion-Related Validity – Kaitlin Henning
A bivariate correlation was used to determine the relationship between nCog scores and college GPA. The results indicated a moderate, statistically significant positive correlation between nCog scores and college GPA (r = .327, p < .01). A linear regression further revealed that approximately 10% of the variance in college GPA was explained by nCog scores (R2 = .107). While these validity tests found a significant relationship between nCog scores and college GPA, it is not as strong as we hoped to see. The nCog scores only explain 10% of the variance in college GPA, suggesting that nCog scores alone should not be used to predict college GPA.
A second bivariate correlation was used to determine the relationship between nCog scores and student satisfaction. The results indicated a weak, but statistically significant positive correlation between nCog scores and Student Satisfaction (r = .205, p < .01). A linear regression further revealed that approximately 4% of the variance in student satisfaction was explained by nCog scores (R2 = .042). These validity tests found a significant relationship between nCog scores and student satisfaction, but a relatively weak one. In this case, the nCog scores explain a mere 4% of the variance in student satisfaction, suggesting that nCog score is not a good predictor of student satisfaction by itself.
Multiple regression analysis was used to determine whether nCog scores explain a significant amount of variance in College GPA above and beyond high school GPA and ACT scores. The results of the regression indicated that nCog scores explained 5.6% more variance than the other predictors. Together, all three predictors explained 23.5% of the variance (R2 =.235, F(1,290) = 29.7, p <. 01). In comparison, nCog scores are better at explaining differences in college GPA than the ACT and high school GPA are. However, using all three of these measures together appears to be much better at explaining college GPA than using nCog scores alone.
Another multiple regression analysis was conducted to decide whether nCog scores explain a significant amount of variance in student satisfaction above and beyond high school GPA and ACT scores. The results of the regression indicated that nCog scores explained 3.8% more variance than the other predictors. Together, all three predictors explained 4.4% of the variance (R2 =.044,F(1,290) = 4.43, p < .01). In this case, the nCog scores are better than high school GPA and ACT scores in explaining college student satisfaction. As mentioned before the nCog scores are not a good predictor of student satisfaction alone, nor do they appear to improve when taking high school GPA and ACT scores into account.
this is a link to explain insurctions
and this is the spss date file