Research Methods

marmaydec
Unit6.docx

Running Header: HYPOTHESIS TESTING 1

HYPOTHESIS TESTING 6

Hypothesis Testing

Insert Your Name Here

Insert University Here

Course Name Here

Instructor Name Here

Date

Data Analysis: Hypothesis Testing

This study uses Sun Coast Remediation data set to conduct an independent samples t test, dependent samples (paired samples) t test, and ANOVA test. Independent samples t test is suitable when the 2 variables whose means are to be compared do not depend on each other. Paired samples t test is suitable if the variable is measured in 2 periods to examine change. ANOVA is used when the mean is compared for more than 2 variables.

Independent Samples t Test: Hypothesis Testing

Independent sample t test was used to examine the difference in means for Group A prior to Training Scores and Group B Revised Training Scores. The null and alternative hypotheses are as shown below:

Ho1: There is no statistically significant difference in mean values between Group A prior to Training Scores and Group B Revised Training Scores.

Ha1: There is a statistically significant difference in mean values between Group A prior to Training Scores and Group B Revised Training Scores.

Data is analyzed using Excel Data Analysis Toolpak and the results are as shown in table 1.

Table 1: Independent Sample t test

According to the analysis, Group B Revised Training Scores had a higher mean of 84.77 compared to the mean of Group A Prior Training Scores which was 69.79. The t-statistics t(87) = -9.67, p = 1.93983E-15 which is less than .05 (Warner, 2020). This implies that we reject the null hypothesis and conclude that there is a statistically significant difference in mean values between Group A prior to Training Scores and Group B Revised Training Scores.

Dependent Samples (Paired Samples) t Test: Hypothesis Testing

Paired samples t test was used to examine the difference in means for employees’ blood levels before exposure to lead and the same employees’ blood levels after exposure to lead. In this case, the variables used are Pre-Exposure μg/dL and Post-Exposure μg/dL. The null and alternative hypotheses will be given as follows:

Ho2: There is no statistically significant difference in mean values between employees’ Pre-Exposure μg/dL and Post-Exposure μg/dL.

Ha2: There is a statistically significant difference in mean values between employees’ Pre-Exposure μg/dL and Post-Exposure μg/dL.

The output for the analysis is as shown in table 2 below;

Table 2: Paired sample t test

The results shows that the mean for employees’ Post-Exposure μg/dL was slightly higher (33.29) compared to mean for employees’ Pre-Exposure μg/dL which was 32.86 μg/dL. The t test statistics t (48) = -1.93, p = 0.06 which is greater than 0.05 implying we fail to reject the null hypothesis and conclude that there is no statistically significant difference in mean values between employees’ Pre-Exposure μg/dL and Post-Exposure μg/dL (Ross, & Willson, 2017).

ANOVA: Hypothesis Testing

One-Way ANOVA was used to examine the difference in means for 4 consulting project return on investments in percentages. The 4 projects are A=Air, B=Soil, C=Water and D=Training. The null and alternative hypotheses will be given as follows:

Ho3: There is no statistically significant difference in mean return on investments between Air, Soil, Water and Training projects.

Ha3: There is a statistically significant difference in mean return on investments between Air, Soil, Water and Training projects.

The output is as shown in table 3.

Table 3: One-Way ANOVA

The results shows that the mean of return on investment for the soil project was higher (9.1%), followed by the mean of return on investment for air project (8.9%), then water project (7%) and the least was training project (5.4%).

The ANOVA results shows that differences were statistically significant where F (3, 76) = 11.92, p = 1.76E-06 which is less than 0.05 implying that we reject the null hypothesis and conclude that there is a statistically significant difference in mean return on investments between Air, Soil, Water and Training projects (Kim, 2017).

References

Warner, R. M. (2020). Applied statistics II: Multivariable and multivariate techniques. SAGE Publications, Incorporated.

Ross, A., & Willson, V. L. (2017). Paired samples T-test. In Basic and advanced statistical tests (pp. 17-19). Brill Sense.

Kim, T. K. (2017). Understanding one-way ANOVA using conceptual figures. Korean journal of anesthesiology70(1), 22.

t-Test: Two-Sample Assuming Unequal Variances

Group A Prior

Training Scores

Group B Revised

Training Scores

Mean69.7903225884.77419355

Variance122.00449526.96456901

Observations6262

Hypothesized Mean Difference0

df87

t Stat-9.666557191

P(T<=t) one-tail9.69914E-16

t Critical one-tail1.662557349

P(T<=t) two-tail1.93983E-15

t Critical two-tail1.987608282

t-Test: Paired Two Sample for Means

Pre-Exposure

μg/dL

Post-Exposure

μg/dL

Mean32.8571428633.28571429

Variance150.4583333155.5

Observations4949

Pearson Correlation0.992236043

Hypothesized Mean Difference0

df48

t Stat-1.929802563

P(T<=t) one-tail0.029776357

t Critical one-tail1.677224196

P(T<=t) two-tail0.059552714

t Critical two-tail2.010634758

Anova: Single Factor

SUMMARY

GroupsCountSumAverageVariance

A = Air201788.99.357895

B = Soil201829.13.042105

C = Water2014076.631579

D = Training201085.41.410526

ANOVA

Source of VariationSSdfMSFP-valueF crit

Between Groups182.8360.9333311.92311.76E-062.724944

Within Groups388.4765.110526

Total571.279