Complete the activities, answer the questions, and save your work in this document. SHOW AS MANY STEPS TO YOUR WORK AS POSSIBLE! The final document must be posted online through the assignments page in Blackboard. Save the document with your name and "_Assignment_Three" in the file name (ex. John_Doe_Assignment_Three.doc). Turn in the document through the assignments page in Blackboard.

This statistics lab is focused on association, correlation and regression. You will need to be prepared for the following statistical computations: chi-square, gamma, scatterplots, Pearson’s r, r ² and ordinary least squares (OLS) regression (aka linear regression).

In order to do this type of analysis, you need to be prepared with the appropriate type of variables. From your readings, you should know that correlation and linear regression are based on the assumption that you are working with interval-ratio level variables. As previously noted in this class, in the social and psychological sciences we are more likely to have access to ordinal level variables than interval-ratio level variables. In attitudinal research, however, it has become common to combine multiple ordinal level variables together to create a new variable that is then treated statistically as an interval-level variable. Luckily, we already recoded variables and created a summated-rating scale measuring trust in local institutions as part of Assignment Two. You will need to use the summated-rating scale again for this assignment.

Gamma – Hand/Calculator Computations

Researchers have shown a link between stress and health. Some studies demonstrate that experiencing financial stress has a negative influence on health. Although the 2007 Delta Rural Poll did not include a question about financial stress, it did include an item focused on level of financial satisfaction. Use the information in the following table to test the theoretical argument that financial satisfaction influences health.

Cross tabulation of self-rated health by financial satisfaction (2007 DRP)

1. State the null hypothesis for the association between financial satisfaction and self-rated health.
2. Find the chi-square critical value with an alpha of 0.05.
3. Calculate the chi-square observed value.
4. Determine whether you reject or fail to reject the null hypothesis.
5. Using the data in the table, calculate column percent values and construct a new table. Write a sentence describing the pattern.
6. Calculate the gamma value. Write a sentence describing the strength and the direction of the association.
   Gamma – SPSS Computations
   Good science requires replication. So, now you want to revisit the association between financial satisfaction (q_58) and self-rated health (rec_q45) using 2009 Delta Rural Poll data.
   NOTE – REMEMBER TO USE THE RECODED SELF-RATED HEALTH VARIABLE WITH ORDINAL CATEGORIES RANGING FROM 1 = Poor to 4 = EXCELLENT. THIS SHOULD BE THE VARIABLE THAT YOU RECODED FOR ASSIGNMENT TWO. IF YOU DID NOT DO THIS FOR ASSIGNMENT TWO, THEN YOU NEED TO DO IT FOR THIS ASSIGNMENT.
   Here are instructions for generating the crosstabulation, obtaining column percents and calculating test statistics (chi-square, gamma).
   Analyze Descriptive Statistics Crosstabs
   Choose rec_q45 (or whatever you named the recoded health variable) for the row variable.
   Choose q_58 for the column variable.
   Click on the Statistics button, check the boxes for chi-square and gamma, and click Continue.
   Click on the Cells button, check the boxes for Observed Count and Column Percentages, and click Continue.
   Click OK.
   (Note that SPSS also generates a sig. value for gamma. This calculation is based on a t-test and tells whether you can reject the null hypothesis that gamma is equal to zero. As with chi-square, if this sig. value is less than the alpha of 0.05, then you can reject the null hypothesis.)
7. State the null hypothesis for the association between financial satisfaction and self-rated health.
8. Use SPSS to calculate the chi-square observed value. Insert the output tables here.
9. Use the SPSS output tables to determine whether you reject or fail to reject the null hypothesis.
10. Use SPSS to calculate column percent values. Insert the SPSS output table here and write a sentence describing the pattern.
11. Use SPSS to calculate the gamma value. Insert the SPSS output table here and write a sentence describing the strength and direction of the association.
    Spearman’s rho and Pearson’s r – Hand/Calculator Computations
    A researcher wants to explore the relationship between age and trust in local institutions (a summated-rating scale). As a researcher assistant, you will use a sub-sample of data from the 2009 Delta Rural Poll for this analysis.

    Respondent	Age	Trust in Local Institutions
    A	23	2.00
    B	24	2.00
    C	36	3.50
    D	42	4.00
    E	35	4.00
    F	35	3.00
    G	43	3.25
    H	48	3.50
    I	54	3.50
    J	52	3.25
    K	49	3.00
    L	52	4.00
    M	49	2.75
    N	50	3.00
    O	62	2.00
    P	58	3.50
    Q	60	2.75
    R	65	3.00
    S	81	3.50
    T	76	3.00

12. Calculate Spearman’s rho. What is this value? Write a sentence interpreting the strength and direction of the association.
    14. Write a sentence accounting for the difference in values between Spearman’s rho and Pearson’s r.
    Scatterplots and Pearson’s r – SPSS Computations
    The previous analysis was conducted using data from a relatively small sub-sample of respondents. Now, you are going to expand your research to the full sample from the 2009 Delta Rural Poll.
    15. Create a scatterplot to visualize the relationship between age and trust in local institutions.
    Graphs Scatter Simple
    Select inst_trust for the variable in the Y axis
    Select q_53 (make sure to use the original age variable, not the recoded age) for the variable in the X axis
    OK
    1. Copy and paste the scatterplot here.
    2. Write a sentence about any patterns that you can identify.

16. Calculate Pearson’s r.

Analyze Correlate Bivariate

Select inst_trust for the variables window AND select q_53 (make sure to use the original age variable, not the recoded age) for the variables window

Make sure that the Pearson box is checked

OK

1. Copy and paste the SPSS output here.
2. Write a sentence about the correlation describing the strength and direction of the relationship.
3. Based on the results, would you say the results are statistically significant? How did you decide?

Now you really want to show off your research skills. You are going to conduct linear regression to determine the influence of the independent variable age on the dependent variable trust in local institutions.

17. State the null hypothesis for the relationship between age and trust in local institutions.

18. Conduct regression analysis.

Analyze Regression Linear

Select inst_trust for the dependent window

Select q_53 (make sure to use the original age variable, not the recoded age) for the independent window

OK

1. Copy and paste the SPSS output here.
2. Based on the results, how much of the variation in trust in local institutions is explained by the model including the variable age?
3. Is the overall regression model statistically significant? How do you know?
4. In standardized terms, what is the relationship between age and trust in local institutions? Is this statistically significant? How do you know?
5. Using the results from this problem, write the regression line formula with the appropriate values (use the unstandardized coefficients in the formula).
6. Using the formula, calculate the predicted trust in local institutions score for a respondent who is 35 years old. What is the predicted score?
7. Using the formula, calculate the predicted trust in local institutions score for a respondent who is 65 years old. What is the predicted score?

19. THINK LIKE A RESEARCHER!!! - From your analyses thus far, age does not appear to be a particularly useful variable for explaining variation in people’s trust in local institutions. In essay form, discuss other variables in the Delta Rural Poll dataset you think would be helpful. Provide justification for your argument.

Price: $37.06

Solution: The downloadable solution consists of 18 pages, 1906 words and 31 charts.
Deliverable: Word Document