In: Statistics and Probability
Problem Set 2: Linear Regression Analysis
Research Scenario: A social psychologist is interested in whether the number of days spent in a refugee camp predicts trauma levels in recently resettled refugees. He interviews 17 refugees to determine how many days they spent in a refugee camp before being resettled, then administers the Harvard Trauma Questionnaire Part IV (HTQ Part 4), where a higher score indicates higher levels of trauma (Mollica et al., 1992). He compiles the information in the table below.
Using this table, enter the data into a new SPSS data file and run a linear regression analysis to test whether days in a refugee camp predict HTQ-4 scores. Create a scatterplot with a regression line to show the relationship between the variables.
Days Spent in Refugee Camp |
HTQ Part 4 Score |
12 |
0.4 |
73 |
1.1 |
60 |
0.9 |
105 |
2.3 |
98 |
1.7 |
76 |
0.3 |
89 |
0.7 |
173 |
2.6 |
189 |
3.1 |
203 |
3.0 |
138 |
1.9 |
215 |
2.5 |
71 |
0.7 |
67 |
1.2 |
63 |
1.8 |
184 |
2.9 |
63 |
0.6 |
Entering the given data and running a simple linear regression, using SPSS,
We get the output:
Interpretation and Results:
A simple linear regression was run to predict 'HTQ' Score' using 'No. of days spent in a refugee camp'. A significant regression equation was obtained F(1,15) = 52.383, p < 0.001), with an R2 of 0.777.The predicted HTQ score can be computed using the fitted equation:
Predicted HTQ score = 0.079 + 0.014 (Days spent in Refugee camp)
From the estimated slope coefficient, we find that the HTQ score increased by 0.014 units for every additional day spent at the refugee camp.
To obtain a scatter plot for the data:
We get the output:
To display the fitted regression line, activate the output:
We get the output: