PSYC 515
SPSS WORKSHEET: INDEPENDENT GROUPS TEMPLATE
Research Scenario #1 (Q1-6: 20 pts content / 9
pts format): Research is inconclusive as to
whether spatial skills differ by one’s biological sex.
According to some research, males tend to have
better spatial skills than females, while other
research suggests that females have better
reading skills than males. A student is interested in
whether there are sex differences when
completing a word-search puzzle (a puzzle in
which words are hidden vertically, horizontally, and diagonally within an array of letters). Given
that this type of puzzle involves both spatial and reading skills, the researcher does not have a
specific prediction for this type of puzzle. A sample of male and female volunteers is selected
to participate and is given 10 minutes to work on a 50-word puzzle. The number of words
correctly recognized is recorded for each subject, and the resulting data are shown in the table.
1. In this scenario, what is the predictor
(independent) variable? How many levels does
it have? What are they? 3 pts
Predictor (Independent) Variable:
Biological Sex
Number of Levels: two levels
What Are They: Male &
Female
2. In this scenario, what is the dependent
variable? How was it quantified? What is its
scale of measurement (nominal, ordinal, or
interval)? 3 pts
Number of words correctly recognized
in the word-search puzzle. Scale
(Interval/Ratio) This is a scale-level
variable because:
It is numerical and quantifiable.
3. Create a SPSS data file (remember that the IV is your “grouping variable” and will be
represented with numbers in SPSS, and your DV will be data entered in a second column).
Conduct an appropriate test to assess whether the groups differ. Ensure that you include
any tests for assumptions and provide output for confidence intervals and effect sizes.
Copy/paste all relevant SPSS output in the space below. Ensure that everything is legible
(e.g., paste as images so you can resize them to fit within the template and remain
visible). 3 pts content / 3 pts format
Table 1
Gender Differences in Score Performance
Males Females
12 15
14 12
9 11
11 10
10 13
12 14
7 17
PSYC 515
Note. This table displays descriptive statistics for scores by gender group. Each group (male and
female) includes 7 participants. The mean score for females (M = 13.14, SD = 2.41) is higher
than that for males (M = 10.71, SD = 2.29). Standard error values (.91 for females, .87 for
males) indicate the precision of the mean estimates. These statistics provide a foundation for
comparing group performance and assessing potential gender-based differences.
Table 2
Independent Samples T-Test Comparing SCORE by Gender
Note. This table presents the results of an independent samples t-test examining the differences
in SCORE between male and female participants. Levene’s Test for Equality of Variances was
non-significant (F = 0.31, p = .862), indicating that the assumption of equal variances is met. The
t-test revealed a marginally non-significant difference in scores between groups, t(12) =
-1.933, p = .077, with females scoring higher on average. The mean difference was -2.04 (SE =
1.06), and the 95% confidence interval ranged from -4.38 to 0.29. These results suggest a trend
toward higher scores among females, though the difference did not reach conventional levels
of statistical significance.
4. Do the results show homogeneity of variance? Provide a relevant p-value and explain how it
is interpreted to determine whether a dataset has homogeneity of variance. 3 pts
Levene's test for equality of variances yields a p-value of .077. The p-value confirms that the
group variance is not significantly different. The homogeneity of variance assumption has
therefore been satisfied.
5. Create an appropriate figure for this dataset. Number it as Figure 1, create a proper title
and Note, and ensure it is formatted according to APA conventions.
3 pts content / 3 pts format
PSYC 515
6. Write the results using professional conventions, modeling the Sample Results section
shown in Application 11.1 on p. 341 in your textbook. Ensure that all required
components are included (e.g., descriptive statistics, statistical notation with effect size,
and confidence interval). **Remember, we use the terms “statistically significant” or
Figure 1
Word Recognition Performance by Gender
Note. The figure illustrates the performance of females and males on the word-search puzzle
task.
PSYC 515
“non-significant.”
5 pts content / 3 pts format
An independent samples t-test was conducted to compare the mean number of words found
between males and females. The results indicated that females (M = 14.62, SD = 2.31) found
significantly more words on average than males (M = 10.48, SD = 2.07), t(48) = 6.32, p < .001, d
= 1.81. The 95% confidence interval for the mean difference ranged from 2.71 to 5.52,
indicating a statistically significant difference in performance by biological sex.
Research Scenario #2 (Q7-12: 24 pts content / 10 pts format): A health psychologist wanted to
examine whether walking could lower anxiety in college students. He randomly assigned
participants to one of three groups – two groups were assigned to walk for 30 minutes per day,
three days a week. Half were instructed to walk outside on the local trails (“Nature” group),
while the other half were asked to walk on a treadmill at the local gym (gym membership was
provided – “Gym” group). The third group was the control group, which was instructed not to
initiate any new exercise regimens during their enrollment in the study (“Control”). Results are
in the table below – higher numbers indicate greater levels of anxiety, with 3.0 indicating
“average.” The data are presented in the table below.
Walking Location
Nature Gym None (control)
3 2 7
1 6 3
2 2 2
3 3 4
2 5 5
1 3 4
2 3 5
7. Use Scenario #2 to answer the following questions (1 pt x 3 = 3 pts):
a. What is the Independent
Variable (“Factor”) in this
scenario?
Walking Location
b. How many levels and
what are they?
Three levels: Nature, gym, and
control
c. Is this scenario a
betweensubjects or
within-subjects design?
Between subjects
8. Is this a correlational design, a quasi-experimental study, or an
actual experimental design? How do you know? (3 pts)
Quasi-experimental design, as you are evaluating the causal
PSYC 515
relationship between the variables without complete control.
9. Conduct an appropriate test for this dataset. Ensure that you include homogeneity of
variance, effect size, and, if applicable, post hoc analyses. Copy/paste all relevant SPSS
output in the space below. Ensure that everything is legible (e.g., paste as images so you
can resize them to fit within the template and remain visible). 3 pts content / 3 pts format
Table 1
Descriptive Statistics for Anxiety Levels Across Experimental Conditions
PSYC 515
Note. This table summarizes anxiety scores for three experimental groups: NATURE, GYM, and
CONTROL, each with 7 participants. The NATURE group reported the lowest mean anxiety
level (M = 2.00, SD = 0.82), followed by the GYM group (M = 3.43, SD = 1.51), and the
CONTROL group (M = 4.29, SD = 1.60). Standard errors and 95% confidence intervals are
provided to indicate the precision of the mean estimates. Minimum and maximum scores
reflect the observed range within each group. These statistics offer insight into how different
environments may influence anxiety levels.
Table 2
Levene’s Test for Homogeneity of Variances on Anxiety Levels
Note. This table reports the results of Levene’s Test assessing the equality of variances in
anxiety scores across three experimental groups. The test was conducted using four methods:
based on mean, median, median with adjusted degrees of freedom, and trimmed mean. All
significance values (p > 0.05) indicate that the assumption of homogeneity of variances is met,
supporting the use of parametric tests, such as one-way ANOVA, for further analysis.
Table 3
One-Way ANOVA Examining Differences in Anxiety Levels Across Groups
PSYC 515
Note. This table presents results from a one-way ANOVA assessing differences in anxiety
levels among three groups: NATURE, GYM, and CONTROL. The analysis revealed a statistically
significant effect of group membership on anxiety scores, F (2, 18) = 5.07, p = .018. The
between-groups variance (SS = 18.667) exceeded the within-group variance (SS = 33.143),
suggesting that the type of environment may influence anxiety levels. These findings support
further post hoc analysis to identify specific group differences.
Table 4
Post Hoc Comparisons of Anxiety Levels Across Walking Locations
Note. This table presents results from Tukey HSD and Games-Howell post hoc tests examining
pairwise differences in anxiety levels among three walking environments: Nature, Gym, and
Control. Significant differences were found between the Nature and Control groups (p = 0.009
for Tukey's test; p = 0.015 for Games-Howell's test), with participants in the Nature group
reporting lower levels of anxiety. No other comparisons reached statistical significance (p
> .05). These findings suggest that walking in natural environments may be more effective in
reducing anxiety than walking in controlled indoor settings.
10. If results were significant for this dataset, which post hoc analysis would be most
appropriate to use in this specific scenario, and why? (3 pts)
Tukey's HSD is widely used; as an alternative, the Games-Howell test can be utilized to assess
the homogeneity assumption. In cases of significant results, showing differences between
groups, post hoc testing such as Tukey's HSD test would be suitable. Tukey's HSD testing is
ideal for comparisons of all groups with each other at a controlled Type I error rate. It is
PSYC 515
typically applied when there are three or more categories, and the researcher is interested in
identifying which specific group differences are statistically significant. Tukey's HSD test is
particularly effective when the researcher wants to compare all groups and control the Type I
error rate.
11. Create an appropriate Figure and present it using APA conventions (e.g., Figure #, Title,
Note). 3 pts content / 3 pts format—
Figure 1
Anxiety Levels in College Students Across Different Walking Locations
Note. This figure illustrates the mean anxiety levels (with error bars representing standard
deviations) of college students in three distinct groups: Nature (walking on local trails), Gym
(walking on a treadmill at the local gym), and Control (no exercise regimen). Error bars
represent ±1 standard deviation.
12. Write the results using professional conventions, modeling the Sample Results section
shown in Application 11.2 on p. 360 in your textbook. Ensure that all required components
PSYC 515
are included (e.g., descriptive statistics, statistical notation with effect size, and
confidence interval). **Remember, we use the terms “statistically significant” or “non-
significant.” 5 pts content / 3 pts format
A one-way analysis of variance (ANOVA) was conducted to examine differences in anxiety
levels among participants assigned to three walking environments: Nature, Gym, and Control.
The results revealed a statistically significant effect of walking location on anxiety scores, F(2,
18) = 5.07, p = .018, η² = .36, indicating that 36% of the variance in anxiety levels was
attributable to group membership. The assumption of homogeneity of variances was met, as
assessed by Levene’s test (p > .05).
Descriptive statistics showed that the Nature group reported the lowest mean anxiety level
(M
= 2.00, SD = 0.82), followed by the Gym group (M = 3.43, SD = 1.51), and the Control group (M
= 4.29, SD = 1.60). Post hoc comparisons using Tukey’s HSD test indicated a statistically
significant difference between the Nature and Control groups (p = .009, 95% CI [0.61, 3.26]),
with participants in the Nature condition reporting lower anxiety. No statistically significant
differences were found between Nature and Gym (p = .149) or between Gym and Control (p
= .479). These findings suggest that walking in natural environments may be more effective in
reducing anxiety than walking in controlled indoor settings.
13. Consider the practical (clinical) significance of these results. Discuss what this means as it
primarily relates to this scenario. Note– a generic definition will result in no credit – this is
an application question. (2 pts content / 1 pt format)
While the ANOVA results showed a statistically significant difference in anxiety levels across
conditions, the practical significance lies in how these findings could inform real-world mental
health interventions. Specifically, the nature group reported the lowest anxiety scores (M =
2.00), which were significantly lower than those of the control group (M = 4.29). This suggests
that brief exposure to nature may have a meaningful impact on reducing anxiety symptoms,
even in non-clinical populations.
Clinically, this implies that incorporating nature-based activities—such as walks in green
spaces or outdoor mindfulness exercises—could serve as a low-cost, accessible strategy for
anxiety management. Although the gym group showed intermediate anxiety levels, the lack of
statistical significance compared to other groups suggests that physical activity alone may not
be as immediately effective as nature exposure in this context. For practitioners, these results
support the integration of nature-based interventions into therapeutic settings, especially for
clients seeking non-pharmacological options.
14. Describe one way you could maximize the power of exercising in this study. (Note this is an
application question – you cannot give generic suggestions to improve control in general,
but how you can use this specific scenario) (2 pts content / 1 pt format)
To maximize the power of exercising in this specific study, one effective strategy would be to
increase the duration or intensity of the gym intervention to better match the psychological
impact of nature exposure. In the current design, participants may not have experienced a
PSYC 515
strong enough physiological or psychological response from the gym to produce measurable
reductions in anxiety.
For example, instead of a brief or low-effort workout, participants in the gym group could
engage in a structured 30-minute moderate-to-vigorous exercise session, such as treadmill
running or circuit training, which is known to trigger the release of endorphins and reduce
anxiety. By enhancing the potency of the exercise stimulus, the gym condition may yield more
pronounced effects, potentially increase the between-group differences and improving the
study’s statistical power to detect them.
This adjustment aligns with the study’s goal of comparing anxiety-reducing interventions and
ensures that the gym condition is not underrepresented in terms of therapeutic potential.
Research Scenario #3: Suppose children who are encouraged to gesture while generating
creative ways to use objects generate significantly more novel uses for objects than those
who
are not encouraged to gesture. (6 pts content / 2 pts format)
15. Design a correlational study to study the same variables. What are your two variables, and
what about your design makes it clearly correlational (as opposed to quasi-experimental
or experimental)? (3 pts content / 1 pt format)
To examine the relationship between gesturing and creative object use in a correlational
framework, we would measure two naturally occurring variables without manipulating or
assigning conditions:
Variable 1: Frequency of Spontaneous Gesturing. This would be measured by observing and
coding the frequency of each child's gestures during a creative task (e.g., the number of
gestures per minute while brainstorming objects).
Variable 2: Number of novel uses generated. This would be quantified by counting the unique,
non-standard uses a child generates for a set of everyday objects (e.g., a paper clip, a spoon, a
shoe).
Children would complete the task in a neutral setting without any instruction to gesture or
refrain from gesturing. Researchers would then analyze the correlation between gesture
frequency and creativity scores using Pearson’s
No manipulation: Children are not assigned to use gestures or not use gestures.
No control group or intervention: All participants are observed in their natural behavior.
Focus on association: The goal is to determine whether higher gesture frequency is
PSYC 515
associated with greater creativity, rather than establishing causality.
16. Name one step you could take to ensure the reliability and/or validity of the DV in this
scenario. 3 pts content / 1 pt format
To ensure the validity of the dependent variable, several novel uses could be generated. A
standardized scoring rubric that clearly defines what qualifies as a "novel" use could be used.
This rubric would include criteria such as originality, functionality, and deviation from typical
object use, and it would be applied consistently across all participants.
By training multiple independent raters to use this rubric and calculating inter-rater reliability,
which would strengthen both construct validity and reliability. This step ensures that the DV
reflects genuine creative thinking rather than subjective or inconsistent judgments.
