Analyzing the Number of Persons who performed House Keeping Data

Analyzingthe Number of Persons who performed House Keeping

Data

Forthis analysis, I used data from the 2010 General Service Survey(GSS). The respondents were asked several questions that have beenused in thorough analysis. These questions included whether they wereworking full time, whether they were working part-time, whether theywere temporarily not working, retired, unemployed, keeping the house,schooling, and so forth. The first dependent variable is keeping thehouse, which is an interval-ratio measurement (Wilcox,2005). Thenumber reported in this survey is the actual number of people whowere able to keep their houses given different scenarios like whenthey are working full time, working part-time, retired, unemployed,or temporarily not working.

Theindependent variables for the analysis are the number of peopleworking full-time, the number of people working part-time, the numberof people who are temporarily not working and those who did notanswer the survey. The number of people who are working full time isan interval-ratio measurement, and it represents the actual number ofrespondents that reported in the survey to be working full time(Wilcox,2005). Thoseworking part-time is also an interval-ratio interval as it representsthe actual persons who stated in the questionnaire that they wereworking part-time. Those that were temporarily not working are aninterval-ratio independent variable as it represents the actualnumber of people that actually stated that they were temporarily outof employment. This implies that they had been fired, but they werecurrently searching for employment. Lastly, the variable thatmeasured those who did not answer is a nominal variable as the datahas been coded in categories. By this I mean that from the surveythose who did not answer ranged from 0 to 4 meaning that code zeroare those who were working under contract, code 1 are those who werenot under the legal age of employment, code 3 are those who werefreelancing while code 4 are those were earning money throughinvesting. However, in this statistical analysis, I will be zero,that is those working under contract and four, that is those who wereinvestors.

Methods

Inthis research paper, I am conducting two different statisticalanalyses to examine the importance of the four independent variablesthat I stated above. These variables include the number of peopleworking full-time, those who were working part-time at the time thesurvey was conducted, those who were temporarily out of employment,and those who did not answer the questionnaire presented to them onthe dependent variable (housekeeping). First, I examine the averagenumber of people who performed housekeeping and were working undercontract and the average number of people who performed housekeepingand were investors using a two samples hypothesis test (Wilcox,2005). WhenI use this hypothesis test, I will be able to determine whether therewill be a difference between the two means. Lastly, I will computethe measures of central tendency for housekeeping and all the otherindependent variables. This will allow me to determine how theinterval-ratio variables are inter-related.

Results

Toperform my analysis, I must first examine the descriptive analysis ofall the variables that have been used in this research paper.

Table1: Descriptive Statistics

Variables

N

Min

Max

Mean

Median

Mode

SD

Working full time

29437

619

2322

981.233

418.717

Working part-time

6115

107

440

203.833

81.8253

Temporary not working

1253

19

90

41.7667

15.7122

Unemployed

1977

25

148

65.9

33.5861

Retired

8102

144

715

270.067

132.058

School

1841

31

140

61.3667

25.062

Keeping house

9650

199

496

321.667

78.1009

Other

1208

9

155

40.2667

30.2688

No answer

16

0

4

0

Themode is the measure of central tendency of the nominal variable abovethat is ‘no answer’. The mean, on the other hand, is the measureof central tendency for the interval-ratio variables while the medianis the measure of central tendency for the ordinal data, which we dono have in this case. As shown in the table above, on average, thep eople reported as to working full time are 981.233, workingpart-time are 203.833, temporarily not working are 41.7667,unemployed are 65.9, retired are 270.067, attending school are61.3667, house keeping are 321.667, and others are 40.2667. The modeof the nominal variable that is ‘no answer’ is code 0, whichimplies those who were working under contract.

Table2: Two-Sample Hypothesis Test of the number of people performinghousekeeping by no answer

No answer

&nbsp

&nbsp

Under Contract

Investors

&nbsp

&nbsp

&nbsp

Number of People keeping house

15

14.5

N

22

1

N= 23, t= 16.446, p=1

Thetable above shows the mean frequency of the number of people keepinghouses for the investors and those who were under contract. The 22people under contract keep houses for a mean of 15 times while theone person who was an investor kept houses for an average of 14.5times per year.

Althoughthe two variables are different, I conducted a two samples hypothesistest to determine whether the mean difference of the variables isstatistically significant. The test statistics stated that the t–testwas 16.446 while the p-value of one indicated the 5% critical region.Since the p-value of one is larger than the critical value of 0.05, Iam therefore able to conclude that the difference between the meansof the two variables is not statistically significant (Wilcox,2005). I,therefore, fail to reject the null hypothesis that there is nodifference between the two means. I am therefore able to concludethat the average number of people that perform housekeeping who areunder contract is more than that of the investors.

Table3: Correlation Matrix of Number of People who performed Housekeepingas compared o other independent variables

&nbsp

Working full time

Working part-time

Temporary not working

Unemployed

Retired

School

Keeping house

Other

Working full time

1

Working part-time

0.96393

1

Temporary not working

0.85033

0.79207504

1

Unemployed

0.62681

0.720392278

0.56289125

1

Retired

0.9308

0.957356456

0.77868217

0.7399

1

School

0.85222

0.883074944

0.77196701

0.7322

0.89783

1

Keeping house

0.7322

0.000643902

0.23029901

0.04071

-0.0113

0.04418

1

Other

0.86517

0.881928059

0.75063469

0.81396

0.93098

0.84171

0.07176

1

Fromthe table above, I calculated the correlation of all the variables inthe analysis. The number of people performing house keeping has weakpositive correlation with the number of people working part-time(r=0.000644). Similarly, it has a weak positive correlation with thenumber of people schooling (r= 0.0044). Other people also have a weakpositive correlation with the number of people attending school (r=0.0718).

Theoverall statistics presented in table 2 and table three states thatpeople who perform housekeeping will work more on a full-time basisand less on a part-time basis. Fewer of them will also not betemporarily employed fewer of them will be unemployed while retired.Lastly, most people who perform housekeeping will not be attendingschool.

References

Wilcox,R. (2005). Introductionto robust estimation and hypothesis testing.Amsterdam: Elsevier/Academic Press. Retrieved on 24 July 2016.