ACE Practices (Aug 2022)

Scenario A
As a dancer, Amy has experienced the benefits of dancing both physically and mentally. She
notices that she becomes more alert and focused in her work. Therefore, she wonders if dance
practices could be incorporated into school activities to increase students’ alertness. To study
this, she recruited 107 primary school children with approval from their parents. In the first
month, all the children attended 1 dance class per week. In the following month, they attended 3
dance classes per week. At the end of each month, the children had to complete a rapid response
computer game. The longer the amount of seconds taken to complete the computer game, the
poorer the alertness. Amy is confused about her analysis so she runs 2 different analyses and
obtains Output A and Output B as below:

Output A

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Class_Frequency Statistic df Sig. Statistic df Sig.
Alertness One_class .264 107 .000 .757 107 .000
Three_classes .210 107 .000 .863 107 .000
a. Lilliefors Significance Correction

Group Statistics
Class_Frequency N Mean Std. Deviation Std. Error Mean
Alertness One_class 107 117.2897 113.46748 10.96931
Three_classes 107 125.9626 107.45494 10.38806

Independent Samples Test

Levene's Test
for Equality of
Variances t-test for Equality of Means
Sig. 95% Confidence Interval
(2- Mean Std. Error of the Difference
F Sig. t df tailed) Difference Difference Lower Upper

Alertness Equal .517 .473 -.574 212 .567 -8.67290 15.10754 -38.45313 21.10734
variances
assumed

Equal -.574 211.375 .567 -8.67290 15.10754 -38.45364 21.10784

variances
not
assumed
Output B

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Mean_Diff .106 107 .005 .965 107 .007
a. Lilliefors Significance Correction

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

Pair 1 One_class 117.2897 107 113.46748 10.96931

Three_classes 125.9626 107 107.45494 10.38806

Paired Samples Correlations

N Correlation Sig.
Pair 1 One_class & Three_classes 107 -.100 .306

Paired Samples Test

Paired Differences
95% Confidence Interval of
Std. Std. Error the Difference Sig. (2-
Mean Deviation Mean Lower Upper t df tailed)
Pair 1 One_class - -8.67290 163.87575 15.84247 -40.08213 22.73633 -.547 106 .585
Three_classes

1. State the research design.

1 IV 2 levels within-subjects design / 1 IV 2 levels repeated measures design

2. a. State the IV(s) / SV(s)

IV: Frequency of dance classes/ Frequency of dance classes per week/Number of dance
classes

b. What is the scale of measurement for the variables in 2a?

Categorical

3. State the levels of the IV(s) / SV(s).

1 class per week, 3 classes per week; 1 class, 3 classes

4. State the DV(s).

Alertness

5. State the operational definition of the DV(s).

The total time (in seconds) taken to complete a rapid response computer game, whereby
longer time indicates poorer alertness.
6. What could be the research question of this study?
Is there an effect of frequency of dance classes on alertness?

7. Construct a null hypothesis for this study.

There is no effect of frequency of dance classes on alertness.

8. Construct a non-directional alternative hypothesis for this study.

There is an effect of frequency of dance classes on alertness.

9. Construct a directional alternative hypothesis for this study.

Attending 3 dance classes per week will lead to greater alertness in children compared to
attending 1 dance class per week.

10. State the appropriate statistical test to run in order to test the null hypothesis in this study
Repeated measures t-test / Within-subjects t-test / Paired samples t-test.

11. Depending on your answer in Question 10, has the assumption of paired samples correlations/
homogeneity of variances been met? Provide statistical evidence.
The assumption of paired samples correlation between the levels, 1 class and 3 classes was
not met, r(105) = -.10, p = .306. Hence, the results should be interpreted with caution.

12. Can Amy reject the null hypothesis? Report in full and provide statistical evidence.
No, she cannot reject the null hypothesis as there was no significant effect of frequency of
dance classes on alertness, t(106) = -0.55, p =.585, 95% CI [-40.08, 22.74].
Scenario B
Relationships offer social support and past literature indicates that individuals who are embedded
in networks of supportive and helpful others tend to have better physical and emotional health,
and higher levels of life satisfaction. Other than that, social support from others not only has
direct benefits for well-being, but also buffers individuals from the ill effects of stressful life
events, such as unemployment, accidents, illnesses, or periods of low income. This was found to
be especially true when it was a romantic partner such as a spouse. Although the link between
marital status and well-being has been widely replicated, most studies have yet to make
distinctions between unmarried individuals and married individuals on subjective well-being.
With this gap in mind, Dr. Shelby recruited 49 married individuals and 51 unmarried individuals
to participate in his study. Subjective well-being was measured using the BBC Subjective Well-
Being scale (BBC-SWB) where higher average score indicates better subjective well-being.

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Group Statistic df Sig. Statistic df Sig.
SWB
Married .102 49 .200* .976 49 .413
Unmarried .126 51 .042 .970 51 .230

*. This is a lower bound of the true significance.

a. Lilliefors Significance Correction

Group Statistics
Group N Mean Std. Deviation Std. Error Mean

SWB Married 49 3.08163 .66764 .095382

Unmarried 51 2.92647 .20524 .02874

Independent Samples Test

Levene's Test for
Equality of
Variances t-test for Equality of Means
95% Confidence Interval
Sig. (2- Mean Std. Error of the Difference
F Sig. t df tailed) Difference Difference Lower Upper

SWB Equal 47.005 .000 1.584 98 .116 .15516 .09796 -.03924 .34956
variances
assumed
Equal 1.558 56.664 .125 .15516 .09961 -.04434 .35466
variances
not
assumed
1. State the research design.
1 SV 2 levels ex post facto design

2. a. State the IV(s) / SV(s)

SV: Relationship status / Marital status

b. What is the scale of measurement (categorical/continuous) for the variables in 2a?

Categorical

3. State the levels of IV(s) / SV(s).

Married individuals, unmarried individuals

4. State the DV(s).

Subjective well-being

5. State the operational definition of the DV(s).

Average score on the BBC Subjective Well-Being Scale (BBC-SWB), whereby higher
average score indicates greater subjective well-being

6. What could be the research question of this study?

Is there a difference in subjective well-being between married individuals and unmarried
individuals?

7. Construct a null hypothesis for this study.

There is no difference in subjective well-being between married individuals and unmarried
individuals.

8. Construct a non-directional alternative hypothesis for this study.

There is a difference in subjective well-being between married individuals and unmarried
individuals.

9. Based on the information given in the scenario, construct a directional alternative

hypothesis for this study.
Married individuals will have greater subjective well-being than unmarried individuals.

10. State the appropriate statistical test for the research question of this study.
Independent/Between-subjects t-test

11. Is the assumption of normality met? Provide statistical evidence.

Yes, the assumption of normality for subjective well-being was met for married
individuals, Shapiro-Wilk (49) = .98, p = .413, and unmarried individuals, Shapiro-Wilk
(51) = .97, p = .230. Hence, the overall assumption of normality was met.

12. Is the assumption of homogeneity of variances met? Provide statistical

evidence.No, the assumption of homogeneity of variances was not met, F =
47.01, p < .001
13. Does the data support Dr Shelby’s (directional) hypothesis? Can the null hypothesis be
rejected? Provide statistical evidence.
No, the alternative hypothesis cannot be supported, as there was no statistically significant
difference in subjective well-being between married individuals and unmarried individuals,
t(56.66) = 1.56, p = .125, 95% CI [-0.04, 0.35]. Specifically, married individuals did not
have statistically significant higher subjective well-being (M= 3.08, SD= 0.67) as compared
to unmarried individuals (M= 2.93, SD= 0.21). Thus, the null hypothesis cannot be
rejected.
Scenario C
Andre is an avid Mixed Martial Arts fan who has been to many fights. He noticed that there is a
difference in fight performances between amateur fighters and experienced fighters, mainly due
to their difference in endurance. Based on his observations and readings, he noticed that
experienced fighters have better endurance compared to amateur fighters, and he thinks that this
may be attributed to their ability to persevere under stress. Hence, to test this out, he recruited 50
amateur fighters and 50 experienced fighters, and assessed their perseverance under stress. To
measure this, Andre used the Grill's Perseverance Scale, whereby a higher average score
indicates greater ability to persevere under stress. Below is the output from his analysis:

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Skill_level Statistic df Sig. Statistic df Sig.

Perseverance_stress Amateur .109 50 .187 .955 50 .057

Experienced .098 50 .200* .952 50 .043
*. This is a lower bound of the true significance.
a. Lilliefors Significance Correction

Group Statistics
Skill_level N Mean Std. Deviation Std. Error Mean

Perseverance_stress Amateur 50 11.3600 4.77562 .67537

Experienced 50 11.0800 4.70276 .66507

Independent Samples Test

Levene's
Test for
Equality of
Variances t-test for Equality of Means
95% Confidence
Sig. Interval of the
(2- Mean Std. Error Difference
F Sig. t df tailed) Difference Difference Lower Upper
Perseverance_stress Equal .063 .802 .295 98 .768 .28000 .94787 -1.60101 2.16101
variances
assumed

Equal .295 97.977 .768 .28000 .94787 -1.60101 2.16101

variances
not
assumed
1. State the research design.
1 SV 2 levels ex post facto design

2. a. State the IV(s) / SV(s)

SV: Fighter’s skill level

b. What is the scale of measurement (categorical/continuous) for the variables in 2a?

Categorical

3. State the levels of IV(s) / SV(s).

Amateur, Experienced

4. State the DV(s).

Ability to persevere under stress

5. State the operational definition of the DV(s).

Average score on the Grill’s Perseverance Scale, whereby a higher score indicates greater
ability to persevere under stress.

6. What could be the research question of this study?

What is the difference in ability to persevere under stress between amateur fighters and
experienced fighters?

7. Construct a null hypothesis for this study.

There is no difference in ability to persevere under stress between amateur fighters and
experienced fighters.

8. Given the information in the scenario description above, propose an appropriate alternative
hypothesis for Andre.
Experienced fighters will have a greater ability to persevere under stress as compared to
amateur fighters.

9. State the appropriate statistical test to run in order to test the null hypothesis in this study.
Independent t-test

10. Is the assumption of normality met? Provide statistical evidence.

The assumption of normality was met for the ability to persevere under stress of
amateur fighters, Shapiro-Wilk (50) = .96, p = .057, but not met for the ability to
persevere understress of experienced fighters, Shapiro-Wilk (50) = .95, p = .043. Thus,
the overall assumption of normality has been violated.

11. Is the assumption of homogeneity of variances met? Provide statistical evidence.

Yes, Levene’s test showed that the assumption of homogeneity of variances has
beenmet, F= 0.06, p= .802.
12. Is there support for the directional alternative hypothesis? Can the null hypothesis be
rejected? Provide statistical evidence.
No, the directional alternative hypothesis cannot be supported, as experienced fighters (M =
11.08, SD = 4.70) did not have significantly greater ability to persevere under stress as
compared to amateur fighters (M = 11.36, SD = 4.78), t(98) = 0.30, p = .768, 95% CI [-1.60,
2.16]. The null hypothesis cannot be rejected.
Scenario D
Professor Riley is looking for learning strategies to help international students in terms of
English language acquisition. She read that retrieval-based learning provides a more enriching
learning experience as compared to rote learning, that relies heavily on memorization. She
wonders if the retrieval learning approach would result in stronger learning retention than the
rote learning approach. A total of 60 participants were recruited and were taught a list of 18
adjectives over two days of classes. At the end of each class, Group A was assigned in the rote
learning approach condition, where they list down as many adjectives and definitions based on
their recall. Group B was assigned in retrieval learning approach condition, in which they write a
short essay using as many adjectives learnt. After a week, all participants were given a quiz on
the adjectives learnt. Higher total number of correct answers from the quiz indicates stronger
learning retention. Professor Riley’s research assistant produced the 2 sets of output below. Help
Professor Riley make sense of the data.

Output A
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Mean_Diff .105 30 .200* .957 30 .261
*. This is a lower bound of the true significance.
a. Lilliefors Significance Correction

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

Pair 1 Rote_L 8.5667 30 1.67504 .30582

Retrieval_L 11.3333 30 2.79573 .51043

Paired Samples Correlations

N Correlation Sig.
Pair 1 Rote_L & Retrieval_L 30 -.012 .949

Paired Samples Test

Paired Differences
95% Confidence Interval of
Std. Std. Error the Difference Sig. (2-
Mean Deviation Mean Lower Upper t df tailed)
Pair Rote_L - -2.76667 3.27670 .59824 -3.99021 -1.54313 -4.625 29 .000
1 Retrieval_L
Output B
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
L_approach Statistic df Sig. Statistic df Sig.

L_retention Retrieval .098 30 .200* .960 30 .303

Rote .135 30 .169 .947 30 .143
*. This is a lower bound of the true significance.
a. Lilliefors Significance Correction

Group Statistics
L_approach N Mean Std. Deviation Std. Error Mean

L_retention Retrieval 30 11.3333 2.79573 .51043

Rote 30 8.5667 1.67504 .30582

Independent Samples Test

Levene's Test
for Equality of
Variances t-test for Equality of Means
95% Confidence
Sig. Interval of the
(2- Mean Std. Error Difference
F Sig. t df tailed) Difference Difference Lower Upper

L_retention Equal 7.511 .008 4.650 58 .000 2.76667 .59503 1.57558 3.95775
variances
assumed

Equal 4.650 47.444 .000 2.76667 .59503 1.56992 3.96342

variances
not
assumed

1. State the research design.

1 IV 2 levels between-subjects design / 1 IV 2 levels independent design

2. (a) State the IV(s) / SV(s)

Type of learning approach / Learning approach

(b) What is the scale of measurement (categorical/continuous) for the variables in 2a

Categorical

3. State the levels of IV(s) / SV(s).

Rote learning approach, Retrieval learning approach
4. State the DV(s).
Learning retention

5. State the operational definition of the DV(s).

Total number of correct answers on the quiz, whereby higher number indicates better
learning retention.

6. What could be the research question of this study?

Is there an effect of type of learning approach on learning retention?

7. Construct a null hypothesis for this study.

There is no effect of type of learning approach on learning retention.

8. Construct a non-directional alternative hypothesis for this study.

There is an effect of types of learning approaches on learning retention.

9. Construct a directional alternative hypothesis for this study.

Using the rote learning approach will result in poorer learning retention than using the
retrieval learning approach.

10. State the appropriate statistical test to run in order to test the null hypothesis in this study.
Between-subjects t-test/ Independent t-test

11. Is the assumption test of homogeneity of variances met? Provide statistical evidence.
No, Levene’s Test of Equality of Variances showed that the assumption of
homogeneity of variances was not met, F =7.51, p = .008.

12. Is there support for the alternative, non-directional hypothesis? Can the null hypothesis be
rejected? Provide statistical evidence.
Yes, the alternative hypothesis can be supported, as there was a significant effect of type
of learning approaches on learning retention, t(47.44) = 4.65, p < .001, 95% CI [1.57,
3.96]. Thus, the null hypothesis was rejected.
Scenario E
Within the past few decades, there has been a surge of interest in the investigation of
mindfulness as a psychological construct and as a form of clinical intervention. The elements of
mindfulness, namely awareness, openness and nonjudgmental acceptance of one's moment-to-
moment experience, are regarded as potentially effective antidotes against common forms of
psychological distress such as rumination, anxiety, worry, fear, anger, and so on. With this idea
in mind, Dr. Cooper designed a study to examine the effect of mindfulness practice on
psychological well-being. He recruited 90 participants for his study and randomly assigned them
into one of two groups. Participants assigned to the mindfulness practice group were taught the
loving-kindness meditation, while the other group was the control group. At the end of each
session, participants were told to fill in the Psychological Well-Being Scale where higher total
score indicates better psychological well-being.

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Mindfulness Statistic df Sig. Statistic df Sig.
PWB
Mindfulness .128 45 .061 .928 45 .008
No_Mindfulness .137 45 .034 .935 45 .015

a. Lilliefors Significance Correction

Group Statistics
Mindfulness N Mean Std. Deviation Std. Error Mean

PWB
Mindfulness 45 11.13 2.427 .362
No_Mindfulness 45 6.91 3.021 .450

Independent Samples Test

Levene's Test for
Equality of
Variances t-test for Equality of Means
95% Confidence
Interval of the
Sig. (2- Mean Std. Error Difference
F Sig. t df tailed) Difference Difference Lower Upper

PWB Equal variances 1.245 .268 7.308 88 .000 4.222 .578 3.074 5.370
assumed
Equal variances 7.308 84.093 .000 4.222 .578 3.073 5.371
not assumed

1. State the research design.

1 IV 2 levels between-subjects/ 1 IV 2 levels independent design
2. a. State the IV(s) / SV(s)
IV: Mindfulness practice

b. What is the scale of measurement (categorical/continuous) for the variables in 2a?

Categorical

3. State the levels of IV(s) / SV(s).

Mindfulness practice, control group

4. State the DV(s).

Psychological well-being

5. State the operational definition of the DV(s).

Total score on the Psychological Well-Being Scale where higher total score indicates
better psychological well-being

6. What could be the research question of this study?

Is there an effect of mindfulness practice on psychological well-being?

7. Construct a null hypothesis for this study.

There is no effect of mindfulness practice on psychological well-being.

8. Construct a non-directional alternative hypothesis for this study.

There is an effect of mindfulness practice on psychological well-being.

9. Construct a directional alternative hypothesis for this study.

Engaging in mindfulness practice leads to greater psychological well-being than not
engaging in mindfulness practice.

10. State the appropriate statistical test for the research question of this study.
Independent/Between-subjects t-test

11. Is the assumption of normality met? Provide statistical evidence.

No, the assumption of normality for psychological well-being was not met for
mindfulness practice group, Shapiro-Wilk (45) = .93, p = .008, and control group,
Shapiro-Wilk (45) = .94, p = .015. Hence, the overall assumption of normality was
not met.

12. Is the assumption of homogeneity of variances met? Provide statistical

evidence.
Yes, the assumption of homogeneity of variances was met, F = 1.25, p = .268
13. Does the data support Dr. Cooper’s directional hypothesis? Can the null hypothesis be
rejected Provide statistical evidence.
Yes, the alternative hypothesis can be supported, as there was a statistically significant
effect of mindfulness practice on psychological well-being, t(88)= 7.31, p <.001, 95% CI
[3.07, 5.37]. Specifically, the mindfulness practice group (M = 11.13, SD = 2.43) had
statistically significant greater psychological well-being than the non-mindfulness
practice group, (M = 6.91, SD = 3.02). Thus, the null hypothesis can be rejected.
Scenario F
The availability heuristic is a mental bias where people use shortcuts to make judgements about
the likelihood of an event happening based on how easily they can recall an example of that
event. In other words, the more easily you can give an example of an event, the more likely that
will influence your judgement. Based on this concept, Harriet is interested to find out if the
exposure to different types of investment news will influence a person's confidence in
investment. To test this out, she recruits 90 stock brokers as participants. For the first condition,
Harriet assigns the participants as brokers for a fictitious trading firm, while exposing them to
news of successful investments outcomes in recent years. Then, she gives them a hypothetical
capital of RM 5000 and asks them how much they are willing to spend to invest for that
company. A few days later, the same group of participants were called back, and Harriet exposes
them to news of poor investment outcomes, after which they are also required to state how much
they are willing to invest out of the RM 5000 capital. The higher the amount indicates greater
confidence in investment.

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Confidence_investment .074 90 .200* .980 90 .185

*. This is a lower bound of the true significance.

a. Lilliefors Significance Correction

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

Pair 1 Sucessful_news 3887.1462 90 587.54694 61.93288

Poor_news 1972.0782 90 579.25182 61.05850

Paired Samples Correlations

N Correlation Sig.

Pair 1 Sucessful_news & 90 .109 .307

Poor_news

Paired Samples Test

Paired Differences
95% Confidence Interval of
Std. Std. Error the Difference Sig. (2-
Mean Deviation Mean Lower Upper t df tailed)
Pair 1 Sucessful_news 1915.06800 778.85965 82.09902 1751.93900 2078.19700 23.326 89 .000
- Poor_news
1. State the research design.
1 IV 2 levels repeated measures design/ 1 IV 2 levels within-subjects design

2. a. State the IV(s) / SV(s)

Types of investment news

b. What is the scale of measurement (categorical/continuous) for the variables in 2a?

Categorical

3. State the levels of IV(s) / SV(s).

Successful investment news, poor investment news

4. State the DV(s).

Confidence in investment

5. State the operational definition of the DV(s).

Total amount out of the hypothetical RM 5000 capital willing to be invest, whereby the
higher the amount indicates the greater the confidence in investment.

6. What could be the research question of this study?

Is there an effect of types of investment news on confidence in investment?

7. Construct a null hypothesis for this study.

There is no effect of types of investment news on confidence in investment.

8. Construct a non-directional alternative hypothesis for this study.

There is an effect of types of investment news on confidence in investment.

9. Based on the scenario above, construct a directional alternative hypothesis for this study.
Exposure to poor investment news will result in lower confidence in investment compared to
exposure to successful investment news.

10. State the appropriate statistical test for the research question of this study.
Repeated measures t-test

11. Is the assumption of normality met? Provide statistical evidence.

The assumption of normality for confidence in investment has been met, Shapiro-Wilk
(90) = .98, p = .185.

12. Has the assumption of correlations been met?

No, the assumption of correlations between the levels, successful news investment condition
and poor investment news condition was not met, r(88)= .11, p = .307. Hence, the results
should be interpreted with caution.
13. Does the data support Harriet’s non-directional hypothesis? Can the null hypothesis be
rejected? Provide statistical evidence.
Yes, the alternative hypothesis can be supported, as there was a statistically significant
effect of types of investment news on confidence in investment, t(89) = 23.33, p < .001,
95% CI [1751.94, 2078.20]. Hence, the null hypothesis was rejected.