Chi-Square Test

Two categorical variables . Choose two categorical variables that you think may be associated with each other. These could be the variables you used in Lab Assignment 2 (but do not need to be).

1. Two-way table . Present a cross-tabulation of the two categorical variables. If one variable can be viewed as the explanatory variable and the other as the response variable, make the explanatory variable the column variable and present column percentages along with the counts in the table.
2. Bar Chart. Produce a bar chart that represents the counts or percentages of the two-way table.
3. Chi-Square Test . Test the null hypothesis that the two variables are not associated. If the result is significant at the 5% level, describe the nature of the association.

Multiple Regression

Response variable. Your response variable, Y , must be quantitative.

Explanatory variables . Choose two explanatory variables, X ₁ and X ₂ that you believe to be predictive of Y . The first variable, X ₁ , should be quantitative (for instance a measure of socio-economic status). The second variable, X _2, can be either quantitative or dichotomous. If you choose a dichotomous variable, it should be dummy-coded (0, 1).

1. Univariate descriptive statistics. Present a table summarizing descriptive statistics for each of the three variables. For quantitative variables, you should calculate the mean and standard deviation. For dichotomous variables, you should calculate the percentage of people in each group. For consistency with the regression analysis, the descriptive statistics you calculate should only be for people with data on all three variables. (This is called "listwise deletion of missing data)".
2. Scatterplots for pairs of variables. For each pair of quantitative variables, generate a scatterplot. The easiest way to do this if you have more than two quantitative variables is using a scatterplot matrix. Examine the scatterplots for linearity, possible problems with ceiling or floor effects, and possible problems with outliers. In your write-up, you should note anything unusual you see in the graphs.
3. Regression analysis #1. Perform a regression analysis using X ₁ as the only explanatory variable. In your write-up, note whether the slope coefficient for SES is statistically significant. Also note the magnitude of R ² .
4. Regression analysis #2. Perform a second regression analysis by adding your second predictor variable to the model. (Thus, for this analysis, you’ll have two explanatory variables). Note the size of R ² for this analysis with two predictor variables, and how much R ² has changed . Also note whether each of the slope coefficients (one for each explanatory variable) is statistically significant at the 5% level.

Write a brief (2-3 pages max) summary of your results for this lab. You may structure it in the way that makes most sense to you, but be sure to include any tables, figures, and other output you discuss. Please edit your output – remove anything that you don’t discuss.

Include a do-file of all your analyses in an appendix.

Price: $14.78

Solution: The downloadable solution consists of 7 pages, 778 words and 6 charts.
Deliverable: Word Document