Name: Alec Lo

Course & Section: BS Psych C109

LABORATORY ACTIVITY NO. 4

Hypothesis Testing using a Z Test

Objectives: Perform the steps in hypothesis testing with a Z test and interpret results of the test.

Instructions: Answer each question that follows the given problem. Perform the necessary
calculations and use the Z table provided in class to find critical values. Show your solution where
applicable. (Total: 25 points)

1) A research institute examined the number of mobile devices (smartphones and tablets) in
Filipino households and reported a mean (μ) of 4 devices per household and a standard
deviation (σ) of 2 devices. But I wonder if the statistics in my neighborhood, Blue Hills, would be
different from the national average. Specifically, I have a hunch that households in Blue Hills
might have more mobile devices compared to the national average.

To test this idea, I randomly picked 16 friends of mine living in Blue Hills and asked them how
many smart phones and tablets are in their households. The data set is shown below. (The same
data are also provided in an Excel file if you choose to use Excel for your calculations.)

I would like to perform a Z test to see if the numbers of mobile devices in Blue Hills households
(represented by the sample of 16 households) are significantly higher than the numbers of
mobile devices in the national households. The significance level for my Z test was set at α = .05.
The hypotheses should be directional because I am predicting a specific direction of difference,
and the test should be one-tailed.

Subject # (Household) # of mobile devices

1 6
2 7
3 3
4 6
5 4
6 6
7 5
8 4
9 7
10 4
11 4
12 6
13 5
14 3
15 4
16 6

Page 1 of 4
a. What is the dependent variable in this study? (1 point)
- The dependent variable for this study is the number of mobile devices per household in
Blue Hills.

b. What should be my null and alternative hypotheses? State each hypothesis using both words
and statistical symbol notation. (2 points)
- Alternative Hypothesis: The total number of mobile devices in Chandler’s household are
significantly higher than the number of mobile devices nationwide.
H1:μChandler>μNational
- Null Hypothesis: The total number of mobile devices in Chandler’s household are not
significantly higher than the number of mobile devices nationwide.
H0 : μ Chandler ≤ μ National

c. Calculate the sample mean. (1 point)

- M=ΣX/N = 80/16 = 5

d. Calculate standard error using the population standard deviation σ given in the research
scenario (SE, which is the standard deviation of the sampling distribution). (1 point)
- σ M ¿ σ /√N = 2/√16 = 2/4 = 1/2 = 0.5

e. Calculate the Z statistic (which indicates where our sample mean is located on the sampling
distribution). (1 point)
- μM=μ=4
- σM=0.5
- M=5
z=(M−μM) / σM = (5−4) / 0.5 = 1/ 0.5 = 2

f. Determine the critical Z value(s). Explain how you find the answer. (1 point)
- The critical Z value is 1.64. With a significance level (α) of 0.05, According to the z table, the
nearest values is 5% (0.05 * 100). The potential percentages were 5.05 and 4.95.
Considering it's a one-tailed test, we determined the corresponding z value as 1.64.

g. Compare the Z statistic with the appropriate critical Z value and then draw a conclusion about
the result of the hypothesis test.
- Is the Z statistic more extreme than the critical Z? Illustrate or explain. (1 point)
- Do you “reject” or “fail to reject” the null hypothesis? (1 point)
- We reject the null hypothesis because the z statistic = 2 is more extreme than the critical Z =
1.64.

h. Write a sentence to answer the research question (you can use the wording from the
hypotheses or explain it in another way). (2 points)
- The null hypothesis is rejected because the total number of mobile devices in Chandler’s
household is significantly higher that the number of mobile devices nationwide.

Page 2 of 4
2) According to the CDC report, the mean life expectancy in the US population is currently (µ) 78.5
years with a standard deviation (σ) of 12. A researcher examined the age of death of 100 people
recently recorded in several Arizona hospitals and calculated the mean to be 79.7 years old. He
runs a two-tailed Z test with α = .05 to see if Arizona has a significantly different life expectancy
compared to the US population. Because the researcher is not predicting a direction, the
hypotheses should be non-directional and the test should be two-tailed.

a. What is the dependent variable in this research question? (1 point)

- The dependent variable in this study is the life expectancy in the US population.

b. What should the null and alternative hypotheses? State each hypothesis using both words and
statistical symbol notation. (2 points)
- Alternative Hypothesis: The life expectancy in Arizona is significantly different than the
national life expectancy.
H1:μ Arizona≠μUSpopulation
- Null Hypothesis: The life expectancy in Arizona is not significantly different than the
national life expectancy.
H0:μ Arizona=μUSpopulation

c. Calculate standard error (SE, which is the standard deviation of the sampling distribution). (1
point)
- σM= σ/√N = 12√100 = 12/10 = 1.2 = SE

d. Calculate the Z statistic (which indicates where our sample mean is located on the sampling
distribution) (1 point)
- z = (M−μM)/σM = (79.7−78.5) / 1.2 = 1.2/1.2 = 1
- Z statistic = 1

e. Determine the critical Z value(s). Explain how you find the answer. (1 point)
- The critical Z value is ±1.96. In the case of a two-tailed test, we need to divide the
significance level of 0.05 into two equal parts, resulting in 0.025 for each tail. This division is
indicated by the 2.5% in the tail portion, leading to the critical Z value of ±1.96.

f. Compare the Z statistic with the appropriate critical Z value and then draw a conclusion about
the result of the hypothesis test. What is the answer to the research question now?
- Do you “reject” or “fail to reject” the null hypothesis? Illustrate or explain. (1 point)
- What is the answer to the research question? (1 point)
- There is a failure in rejecting the null hypothesis because the Z statistic (1) is less extreme
than the critical Z value (1.96).

Answer the following questions based on this alternative scenario:

Because a sample of 100 people is small, it may not represent the state of Arizona adequately. So
the researcher decides to continue to collect data until the sample becomes 400. The average life
expectancy remains 79.7, the same as the previous scenario. All other aspects of the study remain
unchanged.

Page 3 of 4
 g. What is the standard error with this sample? (1 point)
- σM= σ√N = 12√400 =12/20 = 0.6 = Standard Error

h. What is the Z statistic with this sample? (1 point)

- Z statistic = (M−μM)/σM = (79.7−78.5) / 0.6 = 1.2/0.6 = 2

i. Compare the Z statistic with the appropriate critical Z value and then draw a conclusion about
the result of the hypothesis test. What is the answer to the research question now?
- Do you “reject” or “fail to reject” the null hypothesis? Illustrate or explain. (1 point)
- What is the answer to the research question? (1 point)
- There is a failure in rejecting the null hypothesis because the Z statistic = 2 is greater than
the critical value = 1.96. Arizona does not have significantly higher life expectancy than the
rest of the US population.

j. Based on the hypothesis test results with the two samples (one with 100 subjects and the other
with 400 subjects), how did the increase in sample size impact the test results in terms of the Z
statistic? (2 points)
- The Z statistic increases by 1 but the standardized effect size remains the same.

Page 4 of 4