RD-Week 3 Assignment
Theresearch question of interest in this research is ‘Do maleemployees earn more than their female employees per month?’ Thisresearch question will enable the study to find out if there is anydifference in the salaries of male employees and female employees.There is contradicting information regarding differences in salariesowing to the gender differences. The statistical test to be used tosolve this research question is the independent samples test. This isbecause there is need to compare the mean salary of two groups. Thegroups are males and females. There is a continuous dependentvariable, which is income/salary and a categorical variable, which isthe gender that has two groups, that is, male and female.
Thenull hypothesis (H0): There is no difference in the salaries of malesand females. The alternative hypothesis (H1): There is a differencein the salaries of males and females. The null hypothesis means thatthere is no difference in the mean salaries between male employeesand female employees, and the alternative hypothesis means that thereis a difference in the mean salaries of females and male employees.
Thereare two types of errors likely to occur, which are type 1-error andtype two errors (Lieberman & Cunningham, 2009). The type oneerror will occur if we reject the null hypothesis while the evidencewe have suggests that we should not have rejected the nullhypothesis. This means that in this case we would have concluded thatthere is a difference in the male employees’ salary and femaleemployees’ salaries while our evidence indicates that there is anydifference in the salary of employees based on their gender. Thesecond type of error that could occur is type two errors whereby wefail to reject the null hypothesis while the evidence we haveindicates that we should have rejected the null hypothesis. Thismeans that we would have concluded that there is no difference in thesalary of female employees and male employees while our evidenceindicates that there is a difference in the salaries based on genderof the employees.
Inthe study, there will be 100 employees, and the sample will compriseof 50 female employees and 50 male employees. The sample willcomprise of employees of different ages so that the sample can berepresentative of the population. The participants will be chosenthrough a simple random sampling method. The research will take intoconsideration the margin of error required, which is 0.05 and theconfidence interval of 95%. Through this method, everybody has a50/50 chance of being selected for the sample size.
Inthe research, there are two variables, which are salary and gender.The variable salary is a continuous, quantitative with an intervalscale of measurement. The variable gender is a discrete, categoricalwith a nominal scale of measurement.
Salaryis the amount of money that an employee receives at the end of themonth for the services offered while gender refers to one beingeither a male or a female. The measurement procedure of the variableis classification into two categories, which are male and female, andwe do count the number of cases in each category. The measurementprocedure for our variable salary is getting the exact amount andusing the addition sign to get our results.
Thestatistical test that will be used for our research the independentsamples t-test, which is normally used to compare means of two groupsof an independent variable on a continuous or discrete dependentvariable (Lowry, 2014). The test is fit for the research questiongiven that the main aim is of the research is to find if there is adifference in the salary of two groups of employees, which is maleand female.
Theinformation that will be got from the results is the descriptivestatistics and the p-value or the F-statistic that are used to drawconclusions. When the p-value is greater than the level ofsignificance, in this case, is 0.05, we do reject the nullhypothesis. The research is also rejecting the null hypothesis whenthe F-statistic is greater than F-critical. The research fails toreject the null hypothesis if the p-value is less than 0.05 or the F-statistic is less than the F-critical (Moore, 2007).
Theassumptions taken by research have to take into consideration is thatthe salaries are normally distributed, and the two groups of thegender should be independent of each other. Another assumption isthat there is an equal variance of salaries in the two groups. Therecould be bias in the sample stage of the research. The conclusionsthat can be drawn from the research is either if gender determinesthe salaries of the employees or not. The conclusion that cannot bedrawn is that only gender leads to inequality in salaries ofemployees. The results of this research can be used to come up withpolicies that can be used to address the differences in salariesbetween male and female employees.
Lieberman,M. D., & Cunningham, W. A. (2009). Type I and Type II errorconcerns in fMRI research: re-balancing the scale. Socialcognitive and affective neuroscience,nsp052.
Lowry,R. (2014). Concepts and applications of inferential statistics.
Moore,D. S. (2007). Thebasic practice of statistics (Vol.2). New York: WH Freeman.