Question

Course assignment is:
Choose a hypothesis, test your hypothesis with an analysis of the data file, and discuss the results by including both descriptive and inferential statistics.

My hypothesis addresses risk-taking in married vs single Firefighters -

H₀ = Null Hypothesis: the amount of risk that married firefighters take is = to the amount of risk that single firefighters take.
H₁ = Alternate Hypothesis: The amount of risk that single firefighters take is > the amount of risk that married firefighters take.

Using SPSS, what would be the best test to test this hypothesis?  

Answer 1

Hypothesis

As defined, it is the proposed explanation for a phenomenon. In this case, the explanation is the 'marital status' of the firefighter and the phenomenon is 'risk taking' in firefighting.

Descriptive statistics

It is called summarizing statistics, which means it just gives a summary of the samples of the test done. It answers the question of what was the single most predominant outcome of the test. It does not focus on what brought about the output.

Inferential statistics

It involves the detailed analysis of the test set. What attributes contributed to the result, how it varies and to what degree it varies among the subjects in the test set. It is on a lower level and is very detailed, unlike descriptive statistics.

SPSS

Statistical Package for the Social Sciences (SPSS), from a psychology standpoint allows the researcher to conduct their own statistical analysis in a science-based observation that includes descriptive and inferential statistical observations. It helps to create a prediction of all possible outcomes of the test carried out.

The test set for the hypothesis

First, a test set has to be ready in order to carry out the testing of the hypothesis.

In this case, a random sample of the firefighter population should be taken up. This data should be obtained with the least amount of factual errors (Is is assumed that the data file is consistent with this data).

Second, is to determine the null and alternate hypothesis.

The null hypothesis will have to be tested first, whether the risk taken by all firefighters, irrespective of their marital status is the same. This should be assigned a mathematical value based on the degree of risk taken among the subjects in the test set. This can be on a percentile, percentile or point basis.

Third, test the alternate hypothesis.

In this case, the hypothesis is that single firefighters take more risk when compared to married firefighters. Just like the previous step, based on the degree of risk, a mathematical value should be assigned as the quantitative threshold.

Steps of hypothesis testing

1. Stating the hypothesis with clarity
2. Planning on how the data should be analyzed
3. Implement the plan
4. Analyse results and accept or reject the null hypothesis

SPSS in this hypothesis

1. Log the participants in rows and the variable in columns
2. Separate participant data by using a nominal variable named 'marital status'
3. Log the threshold values based on the degree of risk
4. Run the data analysis functions
5. Sort or split files if necessary to ensure that data is consistent
6. Re-run analysis
7. Analyze results and come up with the descriptive statistic - Is the answer to the null hypothesis a yes or no? Is it inconsistent?
8. Analyze results and come up with the inferential statistic - Compare the scores between the two runs of test sets (t-test or the analysis of variance between two tests)
9. Come up with the overall results of this test

• If the answer is yes, what factors contributed to it?
• What are the common variables in the data set?
• If the answer is no, what factors contributed to it?
• What are the common variables in the data set?
• If the results are inconsistent - Why?
• What are the major differences among participants in the data set?

Add-On

One more interesting variable that could be included is how many firefighters who have been divorced are part of the test set. This could be used as a variable in the hypothesis that supports that, marriage to a firefighter has it's own pros and cons and the profession itself could be a reason for divorce.