SarahB82 Posted March 30, 2016 Share Posted March 30, 2016 Is anyone really good at them? I am trying to help my brother with his homework due tonight and am falling short. This is the first problem of many. He is struggling so badly in this class and it is online so while there is help it really isn't very specific and definitely not face to face. If anyone could help us figure out what tests to run in SPSS we would be greatful. He said he is thinking Chi-square or Tukey? All foreign languages to me lol. Problem 1: A social psychology class wanted to see if social group size influenced a person’s conformity. To scientifically address this question, they carried out a study in which the instructor randomly assigned 36 non-psychology participants to one of three groups: a group with one other person, a group with 10 people, or a group with 20 people. Each group was asked a series of seven simple questions; the group members were told ahead of time to give incorrect answers. The class observed how the participants responded when the other group members gave blatantly wrong answers. Each person in the social psychology class rated the participants’ conformity level on a scale of 1-10 for each of the seven questions. The class responses were summed and averaged to give each person a conformity score ranging from 7 to 70, with higher scores indicating greater conformity. __________________ Quote Link to comment Share on other sites More sharing options...
SarahB82 Posted March 30, 2016 Author Share Posted March 30, 2016 We are in Alabama- I am afraid no such thing exists here. Thank you for the help though. Quote Link to comment Share on other sites More sharing options...
OH_Homeschooler Posted March 30, 2016 Share Posted March 30, 2016 I'm a little rusty but I think the one-way ANOVA would be the best option. This would allow you to compare the means of the three groups (average amount of conformity of each group, based on the conformity score assigned to each person) to see if there is an overall effect. Then, if the F-value is significant you would do the follow-up Tukey tests to see which groups are significantly different from each other. Quote Link to comment Share on other sites More sharing options...
SarahB82 Posted March 30, 2016 Author Share Posted March 30, 2016 We are in Alabama- I am afraid no such thing exists here. Thank you for the help though. Quote Link to comment Share on other sites More sharing options...
SarahB82 Posted March 30, 2016 Author Share Posted March 30, 2016 I'm a little rusty but I think the one-way ANOVA would be the best option. This would allow you to compare the means of the three groups (average amount of conformity of each group, based on the conformity score assigned to each person) to see if there is an overall effect. Then, if the F-value is significant you would do the follow-up Tukey tests to see which groups are significantly different from each other.He said thank you! And you don't seem rusty at all to me! I took grad level stars but past the t test I don't remember a thing! Quote Link to comment Share on other sites More sharing options...
SarahB82 Posted March 31, 2016 Author Share Posted March 31, 2016 He said he is stuck on this one - which test should he run? He thins an ANOVA with a cross effect of the two predictor variables? Is he on the right track? Problem 9: Administrators at the Graduate School were interested in finding out the effects of two variables: college attended (private [PR] versus public [PU]) and computer literacy (familiar with computers [FA] versus not familiar with computers [NF]) on the GRE scores of applicants to their various graduate programs. They collected data on 24 applicants and saved these in a SPSS dataset called GRE.sav. They had the following alternative hypotheses: Hypothesis 1: applicants who attended private colleges will on average have significantly higher GRE scores than those who attended public universities Hypothesis 2: applicants who are not familiar with computers will on average have significantly lower GRE scores than those who are familiar with computers, and Hypothesis 3: the effect of computer literacy on GRE scores will vary significantly depending on whether one attended a public or a private university. Quote Link to comment Share on other sites More sharing options...
Liz CA Posted March 31, 2016 Share Posted March 31, 2016 (edited) I'm a little rusty but I think the one-way ANOVA would be the best option. This would allow you to compare the means of the three groups (average amount of conformity of each group, based on the conformity score assigned to each person) to see if there is an overall effect. Then, if the F-value is significant you would do the follow-up Tukey tests to see which groups are significantly different from each other. Forgot the original problem already after I read it. Do you have more than two groups to compare? If so, it's usually done with an ANOVA, if only two groups are being compared you can do the independent t- test. Eta: Google. There are some good stats websites around with detailed explanations. Edited March 31, 2016 by Liz CA Quote Link to comment Share on other sites More sharing options...
Liz CA Posted March 31, 2016 Share Posted March 31, 2016 He said he is stuck on this one - which test should he run? He thins an ANOVA with a cross effect of the two predictor variables? Is he on the right track? Problem 9: Administrators at the Graduate School were interested in finding out the effects of two variables: college attended (private [PR] versus public [PU]) and computer literacy (familiar with computers [FA] versus not familiar with computers [NF]) on the GRE scores of applicants to their various graduate programs. They collected data on 24 applicants and saved these in a SPSS dataset called GRE.sav. They had the following alternative hypotheses: Hypothesis 1: applicants who attended private colleges will on average have significantly higher GRE scores than those who attended public universities Hypothesis 2: applicants who are not familiar with computers will on average have significantly lower GRE scores than those who are familiar with computers, and Hypothesis 3: the effect of computer literacy on GRE scores will vary significantly depending on whether one attended a public or a private university. This sounds like a factorial ANOVA with at least two nominal independent variables (which university and computer literacy) and one scale variable - the GRE score. Quote Link to comment Share on other sites More sharing options...
SarahB82 Posted March 31, 2016 Author Share Posted March 31, 2016 I think that is what he did - a 2 way ANOVA. Univariate in SPSS. Since he is gone now I can say he isn't the brightest bulb in the box and if he passes this class it will be only because he has worked extremely hard. It makes me so thankful that things generally come so easy to me. The poor kid has spent hours and hours working on this and will just squeak by if he passes. Also, he is a commisioned officer in the army and has to pass this class to finish his 4 year degree by this summer so he doesn't lose his commission. All that to say, thank you so much for being willing to help! Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.