SPSS 3

Use SPSS to answer the following questions.

1. Imagine a study in which voters were asked to indicate the degree to which they agreed with a series of statements about their beliefs about global warming. For example, one statement was "Global warming is happening and is at least partially caused by human activity." Participants responded using a Likert scale in which 1 = strongly disagree and 7 = strongly agree. After completing the survey, the voters watched a 2-hour film that explained the scientific evidence on the issue. After watching the movie, all the voters answered the same questions again. Scores were combined so that higher scores indicated greater agreement with statements of belief in global warming. THE DATA IS BELOW. The researcher wants to know whether the film has any effect on voters' beliefs about global warming. (25 pts)

Before	After
2	6
5	6
2	2
3	4
4	5
3	2
7	7
2	1
6	7
1	4
3	4
2	1
1	3
4	4
4	4
4	3
2	3
3	1
2	3
4	3
3	6
3	2
5	5
1	2
3	3
5	1
4	7
5	7
3	2
6	7

a) What type of statistical test should you need to run hereand why? (2 pts)

b) Enter the data into SPSS and provideappropriate labels for the variables (e.g., before/after). (4 pts)

c) Write thestatistical null hypothesis in words (talk about the effect of the film)AND then in symbols (i.e., with some type ofm). (2 pts)

d) Highlight the sample means for before and after on the output and label them "d". (1 pt)

e) If you were to graph these means, what does it look like is going on? (1 pt)

f) Highlight the SEM on the output and label it "f". (1 pt)

g) Highlight thet-statistic and label it "g". (1 pt)

h) The researcher chose to do their test as a two-tailed test. Explain why this might be. (2 pts)

i) Highlight thep value (or "sig") that is relevant to this t-statistic and label it "i". (1 pt)

NOTE: if you can't highlight in your screen shot, just include the screen shot and list the means, test statistic, and p value next to the appropriate letters.

j) Based on thisp value, would you say that you have strong evidence against your null hypothesis? Why or why not? (2 pts)

k) Now look at theeffect size that SPSS computed. Find the Cohen's d point estimate. Label it "k". (1 pt)

l) What size effect is this? Small, medium, or large? (1 pt)

m) Now look at the 95% CI around the mean difference that SPSS computed. Highlight it with "m" (1 pt)

n) Briefly explain what this 95% CI tells us (2 pts)

o) Based on the p value, your effect size, and your CI, what would you conclude about the effect of the film? (3 pts)

2. Researchers have long known that people feel lonely when they are ostracized by others. A researcher wonders whether being ostracized by a computer also makes people feel lonely. To test this hypothesis, the researcher recruits students to participate in their study and measures their current feelings of loneliness using the Social Loneliness Scale (SLS). This scale yields scores that are normally distributed and range between 1 and 11 with higher scores indicating more loneliness. After completing this pretest measure, the participants play a game called Cyberball where they play catch with computer-generated characters. The participants know that the "players" in the game are computer-controlled and not controlled by humans. After a series of passing the ball back and forth, the computer suddenly stops passing the ball to the participant. This exclusion (i.e., ostracism) lasts for 20 passes and then the game ends. After the game, all participants complete the SLS again and their scores are compared. The data is below. Use the data and spss to answer the questions below. (25 pts)

pregame	postgame
10	7
5	6
2	2
3	4
4	5
3	2
10	4
2	4
6	8
7	6
3	4
2	1
7	11
4	4
4	4
4	3
2	3
3	5
2	5
4	11
8	7
3	11
5	5
1	2
3	3
5	9
4	7
5	7
3	3
6	3
4	6
2	7

a) This study uses arepeated measures design. What does this mean and how is it different from thematched design, which is often also analyzed with a related-samples t test? (2 pts)

b) Enter the data into SPSS and provideappropriate labels for the variables (e.g., pregame/postgame). (4 pts)

c) Write thestatistical null hypothesis in words (talk about the computer, ostracism, and loneliness)AND then in symbols (i.e., with some type ofm). (2 pts)

d) Highlight thesample means for pregame and postgame on the output and label them "d".

(1 pt)

e) If you were to graph these means, what does it look like is going on? (1 pt)

f) Highlight the SEM on the output and label it "f". (1 pt)

g) Highlight thet-statistic and label it "g". (1 pt)

h) Briefly explain what this t-statistic compares - i.e., what is measured on the top? The bottom? Don't just put the formula in here but tell me conceptually what the t ratio is looking at. Note: your text has a nice discussion of the logic of t-tests that may help here! (2 pts)

i) Assuming a two-tailed test, highlight thep value (or "sig") that is relevant to this t-statistic and label it "i". (1 pt)

NOTE: if you can't highlight in your screen shot, just include the screen shot and list the means, test statistic, and p value next to the appropriate letters.

j) Based on thisp value, would you say that you have strong evidence against your null hypothesis? Why or why not? (2 pts)

k) Now look at theeffect size that SPSS computed. Find the Cohen's d point estimate. Label it "k". (1 pt)

l) What size effect is this? Small, medium, or large? (1 pt)

m) Now look at the 95% CI around the mean difference that SPSS computed. Highlight it with "m" (1 pt)

m) Discuss at least twoassumptionsof this type of t-test. Note: your text will help here! (2 pts)

o) Based on the p value, your effect size, and your CI, what would you conclude about the ostracism by computers and loneliness? (3 pts)