Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 7

Research Methodology

Assignment 1: Independent-Samples t Test

A student researcher wants to know if professors will give more elaborate responses to questions when they receive
an e-mail that begins with a salutation (e.g., “Dear Dr. Smith,”) than when they receive an e-mail that does not begin
with a salutation. They randomly assign 20 professors to receive the exact same e-mail either with or without a
salutation and keep track of the length of the professors’ responses (based on the number of words).

 Run an independent-samples t-test to compare the two groups.

Before conducting the independent t-test, the main assumptions of the analysis were checked. The normality
assumption was supported, as indicated by a Shapiro-Wilk test (W = 0.964, p-value = 0.629). Additionally, a
variance comparison was conducted using an F test, which suggested a violation of the assumption of equal variance
(F = 5.44, p-value = 0.032). Welch’s t-test was used in this case.

Figure 1. The QQPlot Figure 2. The Descriptive Plot

Figure 1 shows that all the points fall approximately along the reference line, thus, we can assume the normality.
Figure 2 demonstrates the descriptive plot of the data.
The results of the t-test showed that there was a significant difference in the mean scores of the two groups.
Specifically, for the group that received emails with a salutation, the mean response length was 21.2 words (SD =
1.75), while for the group without a salutation, the mean response length was 17.4 words (SD = 3.5). The t-value
with 18 degrees of freedom was -3.07, and the p-value was less than 0.01. This indicates that the mean length of
responses in the group with a salutation was significantly higher than in the group without a salutation. The effect
size, was measured using Cohen's d = -1.37.

 Complete the following:


a) State your null hypothesis.
The mean length of professors' responses between the group that received emails with a salutation and the group that
received emails without a salutation is equal.

b) State a directional alternative hypothesis.


The mean length of professors' responses is significantly higher in the group that received emails with a salutation
compared to the group that received emails without a salutation.

c) Identify the IV and the DV of this experiment.


In this study, the independent variable is the presence of a salutation in the email. This variable is represented as a
dummy variable, with a value of 1 indicating the presence of a salutation and 0 indicating the absence of a
salutation. The dependent variable is the length of the professor's response, which is measured in terms of the
number of words.

d) Based on your results, would you reject or retain the null hypothesis? And therefore, what is the exact
probability of making a Type I error? A Type II error?
Based on the results, we would reject the null hypothesis. The p <.01 indicates a statistically significant difference
between the two groups, thus, the probability of making a Type I error (rejecting the null hypothesis when it is true)
is less than the chosen alpha level. The probability of a Type II error (failing to reject the null hypothesis when it is
false) is not provided here.

e) Write up your results as if you were going to include them in a Results section using the correct APA
format (i.e., double-space, italicize statistical notations, etc.).
An independent-samples t-test was conducted to compare the mean length of professors' responses between the

group that received emails with a salutation (M = 21.2, SD = 1.75) and the group that received emails without a

salutation (M = 17.4, SD = 3.5). The results revealed a significant difference in response length between the two

groups (t(18) = -3.07, p = .007, Cohen's d = -1.37). The group with a salutation had significantly longer responses

compared to the group without a salutation.

f) Interpret your results as you would in a Discussion section. Include recommendations for students sending
e-mails to professors and one suggestion for future research.
This brief study offers practical insight not only for students but also for many professionals in different workplace

settings where email communication is prevalent. The results provide evidence that emails with salutations lead to

more elaborate responses from professors. Students seeking detailed feedback or information from their professors

may benefit from including a salutation in the email. Future work is needed to better understand the underlying

factors influencing professors' responses. Also, future studies could explore the effectiveness of salutations in

different contexts, such as inter-employee communication. Additionally, investigating the effect of including a
salutation not only at the beginning but also at the end of emails, on increasing the likelihood of receiving a

comprehensive response can offer valuable insights to researchers and practitioners.


Assignment 2: One-Way ANOVA

In this study, a student conducted an experiment to determine the most pleasant foreign language for her friends and
family to listen to. Participants were randomly assigned to one of four languages and rated their experience on a
scale from 1 to 10.

 Run the appropriate analysis(es) to compare the groups.

Table 1. Descriptive Statistics


Language N M SD SE
Pleasantness German 8 7.25 0.707 0.250
French 8 6.50 0.756 0.267
Spanish 8 6.75 0.707 0.250
Japanese 8 6.88 0.641 0.227

Table 1 includes the descriptive statistics of the data. It can be noticed that the German language had the highest
mean rating, indicating it was perceived as the most pleasant. However, the differences between the mean ratings are
relatively small. To determine if there are significant differences in the average scores of pleasantness across the
four languages, a one-way ANOVA analysis was conducted. Assumptions of normality and homogeneity of
variances were checked.

The Shapiro-Wilk test confirmed that the data for each language level were normally distributed (W = 0.969, p =
0.474). Additionally, Levene's test indicated that the variance across groups was not significantly different (F =
0.333, p = 0.801), supporting the assumption of homogeneity of variances. Considering the assumption of equal
variances, Fisher's LSD test was used. The result of the test revealed a non-significant association between the
variables (p = 0.217), suggesting no strong evidence to reject the null hypothesis of no association.
In summary, the analysis did not find a significant difference in the average scores of pleasantness among the four
languages.

 Complete the following:

a) What is the IV? What are its levels?


In this study, the independent variable is the different languages that participants listened to. This variable is
presented as a factor variable of four different languages which are German, French, Spanish, and Japanese.
b) What’s the DV? What scale of measurement is represented?
The dependent variable is the rating of pleasantness on a scale from 1 to 10. The scale of measurement represented
is an interval scale.

c) State your null hypothesis.


There is no significant difference in the ratings of pleasantness among the four languages.

d) State a directional alternative hypothesis. Be sure you include all levels of the IV.
 The participants rate the student's speech as more pleasant when speaking German compared to French.
 The participants rate the student's speech as more pleasant when speaking German compared to Spanish.
 The participants rate the student's speech as more pleasant when speaking German compared to Japanese.
 The participants rate the student's speech as more pleasant when speaking French compared to Spanish.
 The participants rate the student's speech as more pleasant when speaking French compared to Japanese.
 The participants rate the student's speech as more pleasant when speaking Spanish compared to Japanese.

e) Based on your results, would you reject or retain the null hypothesis? What type of error might you be
making?
Based on the results, the null hypothesis would be retained since the p-value (0.217) is greater than the conventional
significance level (usually 0.05). In this case, we would not reject the null hypothesis. However, there is a possibility
of making a Type II error (false negative) by failing to detect a significant difference when one actually exists.

f) Write up your results as if you were going to include them in a Results section using the correct APA
format (just as it would appear in a research report).
A one-way analysis of variance (ANOVA) was conducted to examine the differences in pleasantness ratings across

four different languages. The assumption of equal variances was met (F = 0.333, p = 0.801). The results of the

ANOVA revealed no significant difference in the average scores of pleasantness among the languages, F(3, 28) =

1.58, p = 0.217. Post-hoc comparisons using the Tukey HSD test showed no significant pairwise differences in the

pleasantness ratings between the languages (p > 0.05). Therefore, the findings indicate that there is no evidence to

support the hypothesis of significant differences in pleasantness ratings among the four languages.

g) Interpret your results as you would in a Discussion section. Do the results support your hypothesis? What
would you recommend to students regarding their choice of languages? If you wanted to improve the
power of the study, what modifications would you recommend?
The findings of the current brief study did not support the hypothesis, by showing that there would not be a

significant difference in pleasantness ratings among the four languages. It can be concluded that there was no

evidence to suggest that one language was significantly more pleasant than the others. Thus, when choosing a

language to learn students should consider other factors such as personal interest, career opportunities, or

accessibility of learning. To improve the power of the study, increasing the sample size would improve the chances

to see whether there is a difference between languages. Though researchers do not have a direct impact on the

strength of the effect, they have an impact on sample size. If the difference in rating between the conditions was

larger, the study would have a greater chance of detecting a true effect, increasing its power. Also, it can be

suggested to use a different measure and control for confounding variables to investigate the impact on pleasantness

to the greater extent.

You might also like