Hypothesis testing
A two sample t-test is a statistical test that explores if there is significance difference between the mean of two given sets of data. For the sample t-test to be efficient there is an assumption that needs to be made; the two populations come from a normal distribution (Rumsey, 2007). To explore whether there is a significant difference between the males and females in terms of the age, the two sample t test results are as shown below.
Hypotheses formulation
Null hypothesis: there is no significant difference in the mean of the age of male and female
Hnull: μmale= μfemale
Alternative hypothesis: there is significant difference in the age of male and female
Hnull: μmale≠ μfemale
Results
For the analysis, we will use the result for the pooled variance since it is assumed that the two samples are generated from a population with equal variance.
Discussion
Conclusion
Test for association
Hypothesis formulation
Null hypothesis: there is no association between current work and work test. This can be also expressed that there is no dependence between the current work status and work test.
Alternative hypothesis: there is association between current work and work test. This can be also expressed that there is dependence between the current work status and work test.
Cross tabulation
This table gives the distribution of the variables of the current work status and the work test. From the contingency table given above, it is therefore appropriate to do the chi-square test. However, the SYSTAT gives a warning as shown below
This is to mean that the expected frequencies in some of the variables are below 5. This can be seen from the frequencies that are recorded as 2 and 0.
Results
The results of the chi-square test are as shown in the table below
Therefore the degree of freedom is calculated as
(4-1)*(3-1) =6
The degree of freedom is very useful in the calculation of the p-value; however, in this case the program calculates the p-value automatically.
Discussion
We reject null hypothesis: there is no association between current work and work test. This can be also expressed that there is no dependence between the current work status and work test. This will lead us to the conclusion that there is association between current work and work test. This is to mean that there is dependence between the current work status and work test.
Conclusion
After the performance of the chi-square test of association, it was found out that there is significant association between the current work and the work test. This is to mean that the work test depends on the current work status. The association can therefore be attributed to factors in the work place. The dependence therefore means that the current work status influences the work test, since work test is the dependant variable.
Sources
Rumsey, D. J. (2007). Intermediate statistics for dummies. Hoboken, NJ: Wiley Pub.
Albright, S. C., Winston, W. L., & Zappe, C. J. (2011). Data analysis and decision making. Mason, Ohio: South-Western/Cengage Learning.
Corder, G. W., & Foreman, D. I. (2009). Nonparametric statistics for non-statisticians: A step-by-step approach. Hoboken, N.J: Wiley.
