Question

1. Now suppose a larger sample (30 pairs vs 10 pairs) was collected and a paired t test was used to analyze the data. The output is shown below.

Paired Samples Statistics
				Mean			N			Std. Deviation				Std. Error Mean
Pair 1		Female salaries		6773.33			30			782.099				142.791
		Male salaries		7213.33			30			875.227				159.794
Paired Samples Correlations
						N			Correlation			Sig.
Pair 1		Female salaries & Male salaries				30			.881			.000
Paired Samples Test
			Paired Differences													t	df		Sig. (2-tailed)
			Mean		Std. Deviation			Std. Error Mean			95% Confidence Interval of the Difference
											Lower		Upper
Pair 1	Female salaries - Male salaries		-440.000		414.063			75.597			-594.614		-285.386			-5.820	29		.000

1. Write a null hypothesis to accompany the test. Assume that you are testing that there are no differences in salaries between genders. Record both a genericversion (through the use symbols) and an Englishversion (using words).

2. Write the results for the hypothesis test. Be sure to include the appropriate statistics.

Answer 1

a) Nuu hypothesis, Ho: uz= ly, i.e, here is no Signiticant differences in salanies betw een genders two Vs) Atternative hypo thesis, H, Llr y too tailed) Le, salenies of two genders ary sgnificanty falary comtssponduny to female s lery comessponding to male whee Nr mean 1 mian

Here, we suppose difference in salaris af lemal &male ', e do = (-12) Lest Stahstic : undler H,the test statis tc is d tcn-i NOLo, ue wlll ude the giren data guesrion in -440.00XJ30 4/4.063 {n=303 -2409.98 (1-0%)4 t21 4 14.063 5.820 at 29 df 5820 Cal value of t, at 29 tab af forom Ne will see the he t table tatcal Cat lenel of Signiticane) there is Reeult aifferences we will auet Ho le genders no in salary between the and,tcalta ,Ho will be rejutd