Week 1 Project – Hypothesis Testing and Inference
This assignment focuses on estimation and hypothesis testing with one-sample and two-sample inferences. The first part you will make tables and not use SPSS. The second part you will use SPSS.
The essence of parametric testing is the use of standard normal distribution tables of probabilities. For each exercise, there will be a sample problem that shows how the calculations are done and at least one problem for you to work out.
For the first assignment, you will not need any statistical software. However, you will use a standardized normal distribution table (a z-score table) provided in the course textbook (Table 3āThe normal distributionāin the Tables section in APPENDIX) to obtain your responses.
ClickĀ hereĀ to access the standardized normal distribution table from your course textbook.
Problem 1: Probability Using Standard Variable z and Normal Distribution Tables
Variables are the things we measure. A hypothesis is a prediction about the relationship between variables. Variables make up the words in a hypothesis.
In the attention-deficit/hyperactivity disorder’s (ADHD’s) hypothetical example provided in the tables below, the research question was: What is the most effective therapy for ADHD? One of the variables isĀ type of therapy.Ā Another variable isĀ change in ADHD-related behavior,Ā given exposure to therapy. You might measure change in the mean seconds of concentration time when children read. This experiment is designed to obtain children’s concentration times while they read a science textbook and to find out whether the therapy used worked on any of the children.
Use the stated µ and Ļ to calculate probabilities of the standard variable z to get the value of p (up to three decimal places). In addition, respond to the following questions for each pair of parameters:
- Which child or children, if any, appeared to come from a significantly different population than the one used in the null hypothesis?
- What happens to the “significance” of each child’s data as the data are progressively more dispersed?
In addition to the above, write a formal statement of conclusion for each child in APA style using the 7th edition of theĀ APA Publication Manual. A report template is provided for submission of your work.
Note: Tables 1 and 2Ā are practice tables with answers.Ā Tables 3 and 4Ā are the assignment tables for you to work on. ClickĀ hereĀ for a template to provide and submit your assignment answers for Tables 3 and 4.Ā
Table 1Ā (µ = 100 seconds and Ļ = 10)
|
Child |
Mean seconds of concentration in an experiment of |
z-score |
p-value |
|
1 |
75 |
-2.50 |
0.0 |
|
2 |
81 |
-1.90 |
0.0 |
|
3 |
89 |
-1.10 |
0.1 |
|
4 |
99 |
-0.10 |
0.4 |
|
5 |
115 |
1.50 |
0.0 |
|
6 |
127 |
2.70 |
0.0 |
|
7 |
138 |
3.80 |
<0.0 |
|
8 |
139 |
3.90 |
<0.0 |
|
9 |
142 |
4.20 |
<0.0 |
|
10 |
148 |
4.80 |
<0.0 |
Table 2Ā (µ = 100 seconds and Ļ = 20)
|
Child |
Mean seconds of concentration in an experiment of |
z-score |
p-value |
|
1 |
75 |
-1.25 |
0.1 |
|
2 |
81 |
-0.95 |
0.1 |
|
3 |
89 |
-0.55 |
0.2 |
|
4 |
99 |
-0.05 |
0.4 |
|
5 |
115 |
0.75 |
0.2 |
|
6 |
127 |
1.35 |
0.0 |
|
7 |
138 |
1.90 |
0.0 |
|
8 |
139 |
1.95 |
0.0 |
|
9 |
142 |
2.10 |
0.0 |
|
10 |
148 |
2.40 |
0.0 |
Table 3Ā (µ = 100 seconds and Ļ = 30)
|
Child |
Mean seconds of concentration in an experiment of |
z-score |
p-value |
|
1 |
75 |
Ā |
|
|
2 |
81 |
Ā |
|
|
3 |
89 |
Ā |
|
|
4 |
99 |
Ā |
|
|
5 |
115 |
Ā |
|
|
6 |
127 |
Ā |
|
|
7 |
138 |
Ā |
|
|
8 |
139 |
Ā |
|
|
9 |
142 |
Ā |
|
|
10 |
148 |
Ā |
Table 4Ā (µ = 100 seconds and Ļ = 40)
|
Child |
Mean seconds of concentration in an experiment of |
z-score |
p-value |
|
1 |
75 |
Ā |
|
|
2 |
81 |
Ā |
|
|
3 |
89 |
Ā |
|
|
4 |
99 |
Ā |
|
|
5 |
115 |
Ā |
|
|
6 |
127 |
Ā |
|
|
7 |
138 |
Ā |
|
|
8 |
139 |
Ā |
|
|
9 |
142 |
Ā |
|
|
10 |
148 |
Ā |
Submission Details:
- Name your documentĀ SU_PHE5020_W1_A3a_LastName_FirstInitial.doc.
- Submit your document to theĀ Submissions AreaĀ byĀ the due date assigned.
Problem 2: Two-Sample Inferences
A two-sample inference deals with dependent and independent inferences. In a two-sample hypothesis testing problem, underlying parameters of two different populations are compared. In a longitudinal (or follow-up) study, the same group of people is followed over time. Two samples are said to be paired when each data point in the first sample is matched and related to a unique data point in the second sample.
This problem demonstrates inference from two dependent (follow-up) samples using the data from the hypothetical study of new cases of tuberculosis (TB) before and after the vaccination was done in several geographical areas in a country in sub-Saharan Africa. Conclusion about the null hypothesis is to note the difference between samples.
The problem that demonstrates inference fromĀ two dependent samplesĀ uses hypothetical data from the TB vaccinations and the number of new cases before and after vaccination.
Table 5: Cases of TB in Different Geographical Regions
|
Geographical regions |
Before vaccination |
After vaccination |
|
1 |
85 |
11 |
|
2 |
77 |
5 |
|
3 |
110 |
14 |
|
4 |
65 |
12 |
|
5 |
81 |
10 |
|
6 |
70 |
7 |
|
7 |
74 |
8 |
|
8 |
84 |
11 |
|
9 |
90 |
9 |
|
10 |
95 |
8 |
First,Ā click hereĀ to install Statistical Package for the Social Sciences (SPSS) Software analytical software package.Ā
ClickĀ hereĀ to access the dataset.
Using SPSS, download the data and perform the analysis, then complete the following:
- Construct a one-sided 95% confidence interval for the true difference in population means.
- Test the null hypothesis that the population means are identical at the 0.05 level of significance.
In addition, in a Microsoft Word document, provide a written interpretation of your results in APA format.
Submission Details:
- Name your SPSS output fileĀ SU_PHE5020_W1_A3b_LastName_FirstInitial.spv.
- Name your documentĀ SU_PHE5020_W1_A3c_LastName_FirstInitial.doc.
- Submit your document to theĀ Submissions AreaĀ byĀ the due date assigned.
Problem 3: Cross-Sectional Study
In a cross-sectional study, the participants are seen at only one point of time. Two samples are said to be independent when the data points in one sample are unrelated to the data points in the second sample.
The problem that demonstrates inference from two independent samples will use hypothetical data from the American Association of Poison Control Centers.
There are two groups of independent data collected in different regions, which also calls for a t-test. The numbers represent the number of recorded cases of poisoning with chemicals in the homes of 100,000 people in two regions.
Table 6: Cases of Poisoning With Chemicals
|
Year |
Region 1 |
Region 2 |
|
1 |
150 |
11 |
|
2 |
160 |
10 |
|
3 |
132 |
14 |
|
4 |
110 |
12 |
|
5 |
85 |
10 |
|
6 |
45 |
11 |
|
7 |
123 |
9 |
|
8 |
180 |
11 |
|
9 |
143 |
10 |
|
10 |
150 |
14 |
ClickĀ hereĀ to access the dataset.
Using SPSS, download the data and perform the analysis, then complete the following:
- Formulate a null and an alternative hypothesis for a two-sided test.
- Conduct the test at the 0.05 level of significance.
In addition, in a Microsoft Word document, provide a written interpretation of your results in APA format.
Submission Details:
- Name your SPSS output fileĀ SU_PHE5020_W1_A3d_LastName_FirstInitial.spv.
- Name your documentĀ SU_PHE5020_W1_A3e_LastName_FirstInitial.doc.
- Submit your document to theĀ Submissions AreaĀ byĀ the due date assigned.
AdditionalĀ Materials
View a pdf transcript ofĀ One-Tailed and Two-Tailed Test (attached)