Statistics - Hypothesis Testing

Academic Writing and Research Skills

D

2016

Testing Hypothesis and Measure of Association

Sample data-portion of the population

1. Test a sample data above using t-test.

T-test is known to be the statistical analysis of mean of two populations (Taeger & Kuhnt, 2014). For instance, a two sample t-test is used to check whether the two given samples are different is basically used when the two normal distributions variances are not known and also when the small sample size is used in an experiment. Therefore, in our case we have used two sample data one ‘performance before training’ and other one ‘performance after training’

Results

2. Test a sample data above using ANOVA

Analysis of Variance (ANOVA) gives statistical test determining whether the means of various groups are equal or not (Lind, Marchal, & Wathen, 2005). Thus, it can be used to generalize the t-test of more than two groups. In addition, ANOVA are helpful since it help in testing (comparing) three or even more means (variables or groups) for the statistical significance. Furthermore, it is mostly utilized to test or determine what impact the independent variables have on the dependent variables in analysis of regression.

Results

3. Test a sample data above using CHI-SQUARE.

A chi-square is a statistical test that is used to compare the data that has been collected with the data expected (Lind, Marchal, & Wathen, 2005). If the chi-square result give large difference, between the expected and collected data, then this will derive a conclusion that there may something causing that significant change. In addition, having significant large differences will help one or permit one to avoid or reject the null hypotheses which shows or gives a conclusion that there is no relationship between two variables (Taeger & Kuhnt, 2014). Therefore, in our case since the significant difference is close to 1 that is, 0.863975804 then we conclude that training the employees have brought a significant change in business operations and performance.

Results

References Lind, D. A., Marchal, W. G., & Wathen, S. A. (2005). Statistical techniques in business & economics. Boston : McGraw-Hill Irwin. Taeger, D., & Kuhnt, S. (2014). Statistical hypthesisi testing with SAS and IR. Chichester, West Sussex: Wiley.

t-Test: Two-Sample Assuming Unequal Variances

Performance before trainingPerfomance after training

Mean53.4545454557.36363636

Variance789.2727273726.2545455

Observations1111

Hypothesized Mean Difference3.909090909

df20

t Test-0.666070548

P(T<=t) one-tail0.256488102

t Critical one-tail1.724718243

P(T<=t) two-tail0.512976204

t Critical two-tail2.085963447