Student Name: Shivanjali Sonone Student Number: 68200195 Section Number: 901

PSYC 218 901 – Lab Assignment 5

t-test for Correlated and Independent Groups

INSTRUCTIONS: This assignment focuses on Student’s t-test for correlated and independent groups. You
should review Chapter 8 of "A Student Guide to SPSS" before trying to complete this assignment. Use
the SPSS file 'Assignment 5 Data.sav' to complete the assignment. This file contains data from the Zhao
and Epley 2022 publication. You should reference the “Assignment 5 experiment overview”, and
publication before starting the analysis.

Learning Objectives:
This assignment is designed with three major learning objectives. To calculate paired samples and
independent samples t-tests on SPSS. To interpret and apply concepts related to t-tests. The third, is to
teach you APA formatting and notation.

APA notation: Proper formatting for statistics is a very important skill in academia. Every manuscript
requires strict formatting for the results section of the paper. You must have good attention to detail,
and remember to properly format your assignment as a last step before you submit. You will be
penalized for failure to use proper formatting. A Student Guide to SPSS is the best resource to consult.
Keep in mind that there are many kinds of formats and not every publication requires APA for their
publications. In addition, your instructors may select a subset of data from the publication, so your
results and/or formatting may not be the same as what you find there.

Marking of lab assignments will be very strict. Use proper APA symbols and notation (e.g. D obt , X 1 −X 2, tobt, p,
d, D, 1 – 2) when reporting your answers. Please pay particular attention to APA guidelines for
reporting p-values. Follow the exact guidelines in your Cuttler SPSS guide (see pages 20, 25-26, 38, and
all the “Reporting the results” sections in Chapter 8 for more details). You will lose marks for failure to
follow these directions.

This lab assignment is due Wednesday April 10th. You must use Canvas to submit your completed
assignment. If you submit your assignment late, 1/8 of the total mark (3.12 out of the total 25
assignment points) will be deducted for each full or partial 24-hour period.

Professor Dudarev 2023 Page 1

DO NOT REPOST OR REDISTRIBUTE ANY PART OF THIS DOCUMENT.
 Student Name: Shivanjali Sonone Student Number: 68200195 Section Number: 901

PSYC 218 901 – Lab Assignment 5

t-test for Correlated and Independent Groups

1. Use the appropriate t-test to analyze the data from the Willingness to help study. Specifically,
evaluate whether people who imagine asking a stranger for help [Requester] perceive the stranger’s
willingness to help as different than someone who imagines being asked for help [Helper]. Use an
alpha level of .05. [10 points]

a. State the null and alternative hypotheses, using general scientific language. (1 point)
- Null Hypothesis (H0): There is no difference in perceived willingness to help between people
who imagine asking a stranger for help and those who imagine being asked for help.
- Alternative Hypothesis (H1): There is a difference in perceived willingness to help between
the two groups.

b. State the null and alternative hypotheses, using specific statistical language. (1 point)
- Null Hypothesis (H0): μ1 = μ2 (where μ1 is the mean perceived willingness to help for the
Requester group and μ2 is the mean perceived willingness to help for the Helper group).
Alternative hypothesis (H1): μ1 ≠ μ2.

c. For this analysis SPSS provides an analysis called Levene’s Test. What does the Levene Test
assess? Briefly describe how to interpret the results of this test and when we would be
concerned with the results (3 points).
- Levene's Test assesses the equality of variances between groups. It determines whether the
variance of the dependent variable is the same across different groups. If the p-value of
Levene's Test is less than the chosen significance level (usually 0.05), it suggests that the
assumption of homogeneity of variances has been violated. This violation is a concern
because it can affect the validity of the results from the t-test, particularly if sample sizes are
unequal.

d. Report the results, using APA style (Confidence intervals and effect size estimates are not
required as part of your APA-style reports for this sub-question). (2 points)

- T(24) = -0.679 , p = 0.504

e. State your decision with respect to H0 and provide a conclusion in general scientific language.
(1 point)
- If the p-value is less than 0.05, we reject the null hypothesis. Therefore, there is not a
statistically significant difference in perceived willingness to help between people who
imagine asking a stranger for help and those who can imagine being asked for help.

Professor Dudarev 2023 Page 2

DO NOT REPOST OR REDISTRIBUTE ANY PART OF THIS DOCUMENT.
 Student Name: Shivanjali Sonone Student Number: 68200195 Section Number: 901

PSYC 218 901 – Lab Assignment 5

t-test for Correlated and Independent Groups
f. Report the effect size estimate, Cohen’s d. Interpret the size and provide an interpretation of
this statistic (2 points)
- Cohen’s d is a measure of effect size that indicates the standardized difference between two
means. It quantifies the magnitude of the difference between groups relative to the
variability within groups. A larger Cohen’s d indicates a larger effect size. Therefore, for this
question, the Cohen’s d is d = 0.490. Hence, the cohen’s d comes under the Medium effect.

2. Use the appropriate t-test to analyze the data from the Willingness to help study. Specifically,
evaluate the hypothesis that the ‘Helpers’ perceive helping to be more inconvenient on the counting
task than the ‘Requesters’ in this experiment. Use an alpha level of .05, 1 tailed. [6 points]

a. State the null and alternative hypotheses, using general scientific language. (1 point)
- Null hypothesis (H0): There is no difference in perceived inconvenience of helping between
Helpers and Requesters.
- Alternative Hypothesis (H1): Helpers perceive helping to be more inconvenient on the
counting task than Requesters.

b. State the null and alternative hypotheses, using specific statistical language. (1 point)
- Null hypothesis (H0): μ1 ≤ μ2 (where μ1 is the mean perceived inconvenience of helping for
Helpers and μ2 is the mean perceived inconvenience of helping for Requesters).
- Alternative hypothesis (H1): μ1 > μ2.

c. Report the results of the appropriate statistical test, using APA notation style (Confidence
intervals and effect size estimates are not required as part of your APA-style reports for this
subquestion). (2 points)

- T(22.551) = -0.674 , p = 0.254

d. State your decision with respect to H0 and provide a conclusion in general scientific language.
(1 point)
- If the p-value is less than 0.05 and the direction of the difference aligns with the alternative
hypothesis (greater inconvenience for Helpers), we reject the null hypothesis. Therefore,
Helpers perceive helping to be more inconvenient on the counting than the Requesters.

e. Report and provide an interpretation of the effect size estimate, Cohen’s d. (1 points)
- Cohen’s d is a measure of effect size that indicates the standardized difference between two
means. It quantifies the magnitude of the difference between groups relative to the
variability within groups. A larger Cohen’s d indicates a larger effect size. Therefore, for this
question, the Cohen’s d is d = 0.490. Hence, the cohen’s d comes under the Medium effect.

Professor Dudarev 2023 Page 3

DO NOT REPOST OR REDISTRIBUTE ANY PART OF THIS DOCUMENT.
 Student Name: Shivanjali Sonone Student Number: 68200195 Section Number: 901

PSYC 218 901 – Lab Assignment 5

t-test for Correlated and Independent Groups

3. Use the appropriate t-test to evaluate whether the act of helping increased positive mood
specifically for those who were assigned to the Helper condition. You will need to use the ‘select
cases’ function to aid with this analysis. Use an alpha level of .05, 1 tailed. [6 points]

a. State the null and alternative hypotheses, using specific statistical language. (1 point)
- Null hypothesis (H0): There is no difference in positive mood before and after helping among
participants in the Helper condition.
- Alternative Hypothesis (H1): Positive mood increases after helping among participants in the
Helper condition.

b. Report the results of the appropriate statistical test, using APA notation style (Confidence
intervals and effect size estimates are not required as part of your APA-style reports for this
subquestion). (2 points)

- T(17) = 0.649 , p = 0.262

c. State your decision with respect to H0 and provide a conclusion in general scientific language.
(1 point)
- Since the p-value is greater than 0.05, we fail to reject the null hypothesis, we have to retain
it. Therefore, there is insufficient evidence to conclude that helping increases positive mood
among participants in the Helper condition.

d. Thinking about order effects, can you identify a potential confound to the interpretation of
the mood data in this study? What could be another conclusion for these data? [2 points]
- One potential confound to the interpretation of the mood data in this study could be order
effects. For example, participants may experience fatigue or boredom as they progress
through the study, regardless of whether they helped or not.
- Another conclusion for these data could be that the act of helping does not necessarily lead
to an immediate increase in positive mood.

4. Given all of the analyses you performed above, draw an overall conclusion for this study. This should
reference all of the analysis you performed. [3 marks]
- Based on the analyses conducted in this study:

1. The analysis comparing the perceived willingness to help between Requesters and Helpers did
not yield a statistically significant difference.

Professor Dudarev 2023 Page 4

DO NOT REPOST OR REDISTRIBUTE ANY PART OF THIS DOCUMENT.
 Student Name: Shivanjali Sonone Student Number: 68200195 Section Number: 901

PSYC 218 901 – Lab Assignment 5

t-test for Correlated and Independent Groups
2. The analysis examining whether Helpers perceive helping to be more inconvenient in the
counting task than Requesters showed a statistically significant result, indicating that Helpers
perceive helping to be significantly more inconvenient.
3. The analysis investigating whether the act of helping increased positive mood specifically for
participants in the Helper condition did not yield a statistically significant difference.

Therefore, considering all these analyses together, we can draw an overall conclusion that while Helpers
perceive helping to be more inconvenient compared to Requesters, there is no evidence to suggest that
the act of helping leads to an immediate increase in positive mood among participants in the Helper
condition. However, it’s important to note potential confounds such as order effects, which could impact
the interpretation of the mood data. Thus, while helping may be perceived as more inconvenient, its
immediate effect on mood may be minimal or influenced by other factors.

Professor Dudarev 2023 Page 5

DO NOT REPOST OR REDISTRIBUTE ANY PART OF THIS DOCUMENT.