Assignment 1 Descriptive Statistics (due 2/4/15-2/9/15) (13. 5 pts)

By SPSS

2. Click Analyze on the tool bar, select Compare Means, and click on Paired-Sample T Test.

3. Highlight both of the column labels for the two data columns (click on one, then click on the second) and click the arrow to move them into the Paired Variables box, and click OK.

4. Annotate the printout, indicating and highlighting all the relevant results, including (5 pts).

a. Average difference score.

b. Where is the t-value? (is it the same as yours, except the rounding error?)

c. Where is the standard error of mean difference? (is it the same as yours?)

d. Where can you find the 95% CI?

e. Is the result significant? What information do you look for in the SPSS output to

find if your result is significant or not?

1. 
   Even a very small effect can be significant if the sample is large enough. For each of the following, determine how large a sample is necessary for the correlation to be significant. Assume a two-tailed test with α = .05 (Note. Because the table does not list every possible df value, you cannot determine every possible sample size. In each case, use the sample size corresponding to the appropriate df value in the table) (1.5).
   

1. 
   A correlation r = .40 __________________
   
2. 
   A correlation r = .30 __________________
   
3. 
   A correlation r = .20 __________________
   

2. 
   A sociologist wanted to see if there was a relationship between a family’s educational status and the eliteness of the college that their oldest child attended. She measured educational status by counting how many years of education beyond high school the parents received. In addition, she measured the eliteness of the school by its yearly tuition, in thousands (e.g., 5 = $5,000). She obtained a random sample of 10 families. The data are shown below.
   

a. Which is a predictor variable (IV) and which is a criterion variable (DV)? (2)

b. Using SPSS, create a scatterplot. Do you see a pattern of a relationship between the two variables? What is the relationship? (positive or negative, strong or weak?) (1)
c. Using SPSS, obtain and list the mean and standard deviation for each variable (1)

d. Calculate by hand the correlation between the number of years post high school education of parents and the eliteness of the college their oldest child attended. Check your answer by using SPSS to calculate this correlation. Attach the SPSS output to your analysis. (7)

e. Calculate r² by hand. What does r² really mean in this context? (1)
f. Determine if the correlation is significant at  = .05 (two-tailed) (4).

State hypotheses only in a statistical form. What is the critical value? What is your decision? What is your conclusion and interpretation?

g. Compute the regression equation for predicting the eliteness of the college (i.e., yearly tuition) from the number of years post high school education of parents by hand. Is your regression equation the same as the one computed by SPSS (except rounding errors)? (3)

h. Using the regression equation, compute by hand the predicted eliteness of college for the first four scores on the list. Show your work. Compare your predicted values to those in your SPSS file. If your hand calculations of the predicted scores values were incorrect, recalculate them (be sure to print out the SPSS data set with the predicted scores and turn them in with the rest of your assignment) (3).

i. Based on the information given, can you conclude that a family’s higher educational status causes the oldest child to attend more elite college? Why or why not? (1)
3. The regression equation is intended to be the “best fitting” straight line for a set of data. What is the criterion for “best fitting”? (1)

4. 
   Briefly explain what is measured by the standard error of estimate (1).
   

4. 
   In general, how is the magnitude of the standard error of estimate related to the value
   

6. By SPSS

Give a label to each of the variables. Click Graphs on the toolbar. Select Legacy/Dialog and Scatter/Dot. Then select Simple scatter, click on Define. Move your Y variable (DV) in the left box into the Y axis box and your X variable (IV) in the left box into the X axis box, and click OK.

a. Click Analyze on the tool bar, select Correlate, and click on Bivariates.

b. One by one move the labels for the two data columns into the Variables box.

c. The Person box should be checked.

d. Click on Options, check Means and standard deviations, click Continue, and then

click OK. This will get you the mean and standard deviation for each variable.

a. Click Analyze on the tool bar, select Regression, and click on Linear.

b. In the left-hand box, highlight the column label for the Y values, then click the arrow

to move the column label into the Dependent Variable box.

c. For one predictor variable, highly the column label for the X values and click the

arrow to move it into the Independent Variable(s) box.

d. Click on Save, check Unstandardized in the predicted values (first box in the upper

left column), click Continue and then OK. SPSS will compute predicted Y scores

and they will appear in your data file (pre 1 in your data file). Print out the output

from the regression analysis as well as your data file with the predicted scores.

a. the mean and standard deviation of each variable.

b. Pearson correlation (is it the same as yours, except the rounding error?)

c. Slope and Y-intercept

d. Print predicted scores (this comes from your data view)
Your name ____________________________
Bonus Question

Spearman Correlation (due 4/20/15) (7 pts)

1. 
   What are the major differences between the Pearson and Spearman correlations? (1)
   

2. 
   Ten patients were ranked for the degree of psychopathology by two clinical psychologists (1 = least pathological, 10 = most pathological). Compute the correlation between two clinical psychologists on ranking for the degree of psychopathology by hand (3).
   

3. 
   Using SPSS, compute the Spearman correlation (select Spearman when you run correlation, not Pearson) (1). Attach SPSS output.
   

4. 
   Did the two psychologists rank the patients in a similar manner? Explain your answer
   

5. 
   If you obtained the Spearman correlation of 1.0, what does this value mean in terms of
   

Your name ____________________________

Do the data indicate that the distribution of accidents for the three age groups is significantly different from the distribution of drivers? Tests with α = .05.

a. State hypotheses in words and in proportions (proportions are only for H₀) (1).

b. Compute the expected frequencies (1).

c. Compute appropriate statistic to test the hypothesis (2).

d. What is your statistical decision? (.5).

e. What is your conclusion and interpretation (do not forget statistical information)? (1).

2. 
   Research suggests that romantic background music increase the likelihood that a woman will give her phone number to a man she has just met (Guéguen & Jacooby, 2010). In the study, women spent time in a waiting room with background music playing. In one condition, the music was a popular love song and for the other condition the music was a neutral song. The participant was then moved to another room in which she was instructed to discuss two food products with a young man. The men were par of the study and were selected because they were rated as average in attractiveness. The experimenter returned to end the study and asked the pair to wait alone for a few minutes. During this time, the man used a scripted line to ask the woman for her phone number. The following table presents data similar to those obtained in the study, showing the number of women who did or did not give their numbers for each music condition.
   

a. By looking at data above, do you see a relationship between type of music and whether women give their phone numbers? If so, what kind of a relationship do you see? (1)

b. State hypotheses in words (2).

c. Compute the expected frequencies (2).

d. Use appropriate statistic to test the hypothesis (2).

e. What is your statistical decision? (.5)

f. Compute the phi-coefficient (2)

g. What is your conclusion and interpretation? (do not forget statistical information)(1)

Your name ____________________________

1. 
   Explain why you should use ANOVA instead of several t tests to evaluate mean differences when an experiment consists of three or more treatment conditions (1).
   

2. 
   New research suggests that watching television, especially medical shows such as Grey’s Anatomy and House can result in more concern about personal health (Ye, 2010). Surveys administered to college students measure television viewing habits and health concerns such as fear of developing the diseases and disorders seen on television. For the following data, students are classified into three categories based on their television viewing patterns and health concerns are measured on a 10-point scale with 1 indicating “none” and 10 indicating “extremely concerned”
   

2. 
   What are your hypotheses? State them both in words and statistical forms (2).
   

c. Conduct a one-way ANOVA by hand (use the worksheet) (9)

d. Compute effect size (eta squared) by hand. What does this mean in this context? (2)

e. Conduct a Tukey test (3).

3. Use SPSS to (3) (Instructions appear on the next page)

a. Perform a one-way ANOVA.

b. Compute means for each level of the IV.

c. Determine which means differ by computing a post hoc test on all pairwise comparisons of the means.

d. Make sure that the results of your computation by hand are the same as those by SPSS, except rounding error (if they are not the same, suspect errors in computations by hand).

4. Annotate the printout, indicating and highlighting all the relevant results (e.g., means, F-value, homogeneity of variance) (1).

5. Using SPS, bar graph the means (1).

6. Write a results section in APA format (please use a word processor). Present the information in the following order (6)

1. 
   State that you used a .05 alpha level for all statistical tests.
   
2. 
   Draw the reader’s attention to your bar graph (i.e., Figure shows or As can be seen in Figure …)
   
3. 
   Present the results of the F test and eta.
   
4. 
   Tell which post hoc test was performed. Then describe the results of the post hoc test. Be sure to include the direction of the difference.
   
5. 
   Based on your findings, what is your recommendation?
   

1. 
   What the purpose for posttests?
   

2. 
   Explain why you would not do posttests if the analysis is comparing only two
   

3. 
   Explain why you would not do posttests if the decision from the ANOVA was to
   

9. Several factors influence the size of the F-ratio. For each of the following, indicate whether it would influence the numerator or the denominator of the F-ratio, and indicate whether the size of the F-ratio would increase or decrease (2).

1. 
   Increase the differences between (among) the sample means.
   

2. 
   Increase the size of the sample variances.