Research Method

Finals Review
Session 3
 Correlation coefficient
(called the Pearson product moment correlation)
• Definition: A description of the overall degree of agreement between the two measures or sets
of scores. It measures the strength and direction of the linear relationship between the two
variables. It ranges from -1 to 1, where -1 indicates a perfect negative correlation (i.e., as one
variable increases, the other decreases), 0 indicates no correlation, and 1 indicates a
perfect positive correlation (i.e., as one variable increases, the other increases as well).
Chi-square
Definition
o The Chi-Square test is a statistical
procedure for determining the difference
between observed and expected data.

o This test can also be used to determine

whether it correlates to the categorical
variables in our data.

o It helps to find out whether a difference

between two categorical variables is due
to chance or a relationship between them.
Chi-square
Goodness-of-fit test (One-way analysis)

Step 1: Line up the observed frequencies.

Step 2: Determine what the appropriate expected frequency is for each observed
frequency (add up the frequencies and divide by the number of cells.

Step 3: Subtract the expected frequency from the observed frequency in each case:

Step 4: Square each of the results of step 3.

Step 5: Divide each of the squared values obtained in step 4 by the expected frequency.

Step 6: To get the observed value of x2 , add the results of step 5.

 Chi-square
Test of Independence (Two-way analysis)

Step 1: Line up the observed frequencies as well as their marginals (row and column totals).

Step 2: Determine what the appropriate expected frequency is for each observed
frequency (multiply the appropriate row and column marginals and divide the result by the
overall total)

Step 3: Subtract the expected frequency from the observed frequency in each case.

Step 4: Square each of the results of step 3.

Step 5: Divide each of the squared values obtained in step 4 by the expected frequency.

Step 6: To get the observed value of x2, add the results of step 5.
 02

Calculations
and Reports
Report problem + purpose
Remember to include the following in your report:
• Set the scene/background to the study
• Gap/Problem
• Research proposal
• Aim(s)
• Research questions /Hypotheses
• Significance
Report problem + purpose
"The Effects of Teacher Feedback on EFL Students'
Writing Performance"
• Research Problem: Providing effective feedback on writing
assignments is a crucial aspect of English language teaching, yet many
teachers struggle to provide feedback that is clear, concise, and helpful
to students. This study aims to investigate the effects of different
types of teacher feedback on EFL students' writing performance, in
order to identify the most effective strategies for providing feedback
that improves students' writing skills.
Report problem + purpose
"The Effects of Teacher Feedback on EFL Students'
Writing Performance"
Research Purpose: The purpose of this study is to examine the effects of different types of
teacher feedback on EFL students' writing performance. Specifically, the study aims to answer
the following research questions:
• What types of teacher feedback (e.g., direct correction, error identification, peer feedback)
are most effective in improving EFL students' writing performance?
• How does the timing and frequency of teacher feedback (e.g., immediate vs. delayed,
frequent vs. infrequent) affect EFL students' writing performance?
• What are EFL students' perceptions of different types of teacher feedback, and how do
these perceptions affect their motivation and engagement in writing tasks?
Report problem + purpose
"The Effects of Teacher Feedback on EFL Students'
Writing Performance"
Conclusion (Optional):
By answering these research questions, the study aims to provide
insights into effective strategies for providing feedback on EFL
students' writing assignments, and to contribute to the ongoing
conversation in the field of English language teaching about the
most effective approaches to teaching writing.
Paired-Samples T Tests
A researcher wanted to investigate whether a new teaching method improved students' listening
comprehension skills compared to a traditional teaching method in ELT. Data were collected and
analyzed with SPSS 27.0. From the SPSS output below, write a report.
Paired-Samples T Tests
Paired-Samples T Tests
• The null hypothesis for the paired-samples t-test is that there is no significant difference
between the mean scores before and after the ELT program. We can use the t-test to
determine whether the observed mean difference of 24.750 is statistically significant.

• Using a two-tailed t-test with 19 degrees of freedom (N-1=20-1=19), and an alpha level
of 0.05, we look up the critical t-value in a t-distribution table or calculator. The critical t-
value is 2.093.

→ The calculated t-value of 26.814 is much greater than the critical t-value of 2.093, which
means we can reject the null hypothesis and conclude that there is a statistically significant
difference between the mean scores before and after the ELT program.
Paired-Samples T Tests
The purpose of this study was to investigate whether a new teaching method improved
students' listening comprehension skills compared to a traditional teaching method in ELT. A paired-
samples t-test was conducted to determine whether there was a significant difference in mean
scores between the two groups. The results showed that the students who received instruction using
the new teaching method had a significantly greater improvement in their listening comprehension skills
than the students who received instruction using the traditional method. Specifically, the mean
difference between the pre-test (M = 61.25, SD = 12.126) and post-test scores (M = 86, SD = 11.192) was
24.75, indicating that there was a substantial improvement in the scores after the program. The standard
deviation of the differences was 4.22, indicating that there was relatively low variability in the extent of
the improvement across students. We also had the calculated t-value at 26.814 (t(19)=26.814, p<0.001,
two-tailed), which was greater than the critical t-value of 2.093 at a significance level of .05 with 19
degrees of freedom. These findings suggest that the new teaching method is highly likely to be more
effective in improving students' listening comprehension skills in ELT.
Paired-Samples T Tests - Practice
A language school implemented a new ELT program for 20 students and wanted to determine
whether the program was effective in improving their English language proficiency. Data were
collected and analyzed with SPSS 27.0. From the SPSS output below, write a report.
Paired-Samples T Tests - Practice
Independent-Samples T Test
A researcher wanted to investigate the effectiveness of two different methods for teaching
English as a second language (ESL) in terms of their impact on the students' scores on a
standardized test. Data were collected and analyzed with SPSS 27.0. From the SPSS output
below, write a report.
Independent-Samples T Test
• If Sig. is lower than 0.05, the variances of the two dependent are not equal;
therefore, the t value in the row Equal variances not assumed is considered.

• If Sig. is higher than or equal to 0.05, the variances of the two groups are equal, so
the t value in the row Equal variances assumed is used.

• If Sig. (2-tailed) is lower than 0.05, the differences between the two groups is
significant, and the null hypothesis can be rejected.

• If Sig. (2-tailed) is higher than or equal to 0.05, the first variable is not significantly
different from the second one, and the null hypothesis cannot be rejected.
Independent-Samples T Test
Independent-Samples T Test
• Sig. = 1 → Is higher than 0.05 → The t value in the row Equal variances assumed
is used → t(48) = 8.66, p<0.01
• Sig. (2-tailed) < .001 → Is lower than 0.05 → The differences between the two
groups are significant, and the null hypothesis can be rejected.
→ The calculated t-value of 8.66 is greater than the critical t-value of 2.021,
indicating that there is a significant difference between the mean scores for the
lecture-based group and the conversation-focused group.
→ Therefore, we reject the null hypothesis and conclude that there is a significant
difference in the effectiveness of the two methods for teaching English as a
second language.
Independent-Samples T Test
The purpose of this study was to investigate the effectiveness of two different methods for
teaching English as a second language (ESL) in terms of their impact on the students' scores on a
standardized test. An independent-samples t-test was carried out and the results showed a
statistically significant difference in the effectiveness of the lecture-based approach and the
conversation-focused approach for teaching English as a second language (t(48) = 8.66, p<.001,
two-tailed). Specifically, the mean scores of the students who received lectured-based
instruction (Mean = 75, SD = 6.124) was significantly lower than that of the conversation-focused
class (Mean = 90, SD = 6.124). With the information above, t(48)= 8.66 and p< .001 , we can
reject the null hypothesis with significance level of over 99.9%, thus suggesting that the
conversation-focused approach maybe more effective in improving students' performance on
the standardized test. However, it's important to note that this study only examined one
particular test and may not generalize to other measures of language proficiency. Further
research could investigate the effectiveness of these two approaches on a broader range of
measures to better understand their relative merits.
Independent-Samples T Test - Practice
A researcher wanted to investigate the effectiveness of two different approaches to
teaching English vocabulary. Data were collected and analyzed with SPSS 27.0. From
the SPSS output below, write a report.
 Independent-Samples T Test - Practice
• If Sig. is lower than 0.05, the variances of the two dependent are not equal;
therefore, the t value in the row Equal variances not assumed is considered.

• If Sig. is higher than or equal to 0.05, the variances of the two groups are equal, so
the t value in the row Equal variances assumed is used.

• If Sig. (2-tailed) is lower than 0.05, the differences between the two groups is
significant, and the null hypothesis can be rejected.

• If Sig. (2-tailed) is higher than or equal to 0.05, the first variable is not significantly
different from the second one, and the null hypothesis cannot be rejected.
Independent-Samples T Test - Practice
Independent-Samples T Test - Practice
• Sig. = -0.919 → Is higher than 0.05 → The t value in the row Equal variances
assumed is used → t(68) = 6.016, p<0.01
• Sig. (2-tailed) < .001 → Is lower than 0.05 → The differences between the two
groups are significant, and the null hypothesis can be rejected.
→ The calculated t-value of 6.016 is greater than the critical t-value of 1.994,
indicating that there is a significant difference in vocabulary learning between
the rote memorization group and the contextualized approach group.
→ Therefore, we reject the null hypothesis and conclude that there is a
significant difference in vocabulary learning between the two approaches.
 Correlation coefficient
A researcher wanted to find out if there is Hours studied Exam score
any correlation between the amount of time
a student spend studying and their 5 75
performance in an exam. Use Pearson 2 60
Product Moment Correlation to check and 3 65
write a report. Remember to show the
process of plugging numbers into the 1 50
formula for r below. 4 70
6 80
7 85
3 60
2 55
4 70
Correlation coefficient
Hours studied Exam score
5 75
2 60
3 65
1 50
4 70
6 80
7 85
3 60
2 55
4 70
Correlation coefficient
Correlation coefficient
• The correlation coefficient r was calculated to be around +0.985
→ a strong positive relationship between the number of hours
students spent studying English outside of class and their scores
on the final exam.

• The value of r = 0.985 is also higher than the critical r-value 0.63
at n = 10 and 𝛼=0.05, meaning the correlation is significance at
the significance level of 0.05.
Correlation coefficient
The purpose of this study was to investigate whether if there is any correlation between the
amount of time a student spend studying and their performance in an exam. The correlation
coefficient r was calculated to be around +0.985, indicating a strong positive relationship
between the number of hours students spent studying English outside of class and their scores
on the final exam. This means that as the number of hours studied increases, the exam scores
tend to increase as well. The value of r = 0.985 is also higher than the critical r-value at n = 10 and
𝛼 = 0.05, meaning the correlation is significance at the significance level of 0.05.

It is important to note that correlation does not imply causation, and there may be other factors
that contribute to students' exam scores besides the number of hours they spend studying
outside of class. Nevertheless, this result suggests that encouraging students to study more
outside of class could be a beneficial strategy for improving their performance on the final exam.
 Correlation coefficient - Practice
Technology use Motivation
A researcher wanted to find out if there is (hours/week) (out of 10)
any correlation between the use of 5 8
technology in language learning and 3 7
students' self-perceived motivation 2 5
towards learning English. Use Pearson 6 9
Product Moment Correlation to check and 4 7
write a report. Remember to show the 7 10
process of plugging numbers into the 8 9
formula for r below. 1 4
5 8
3 6
2 5
6 9
4 7
7 10
8 9
Correlation coefficient - Practice
Technology use Motivation
(hours/week) (out of 10)
5 8
3 7
2 5
6 9
4 7
7 10
8 9
1 4
5 8
3 6
2 5
6 9
4 7
7 10
8 9
Correlation coefficient - Practice
Correlation coefficient - Practice

• The correlation coefficient r was calculated to be around +0.945

→ a strong positive relationship between the use of technology
in language learning and students' self-perceived motivation
towards learning English.

• The value of r = 0.945 is also higher than the critical r-value 0.51
at n = 15 and 𝛼=0.05, meaning the correlation is significance at
the significance level of 0.05.
Chi-square
In a study investigating the relationship between the gender of
students and their preference for learning English through online
classes, 200 students were surveyed and the results are as followed:

Gender Online classes (Yes) Online classes (No)

Male 40 60

Female 50 50

Can we reject the null hypothesis of no difference at an .05 level?

Write a report
Chi-square
Gender Online classes (Yes) Online classes (No) Row marginals (Totals)

Male 40 60

Female 50 50

Column marginals
(Totals)
Chi-square
Chi-square
• The contingency table revealed that there were 40 males and 50 females
who preferred online classes, and 60 males and 50 females who did not
prefer online classes.

• the chi-square statistics is around 2 → Less than the critical value of chi-
square 3.84 for a significance level of 0.05 and 1 degree of freedom is 3.84

→ Fail to reject the null hypothesis of no relationship between gender and

preference for learning English through online classes and conclude that
there is no statistically significant relationship between the two variables.
Chi-square
The study aimed to investigate whether there is a relationship between gender and preference for
learning English through online classes among 200 students. The contingency table revealed that
there were 40 males and 50 females who preferred online classes, and 60 males and 50 females who
did not prefer online classes. After conducting a chi-square test, the chi-square statistics is around 2.
We also have the critical value of chi-square for a significance level of 0.05 and 1 degree of freedom
is 3.84, Since the calculated chi-square statistic is less than the critical value, we fail to reject the null
hypothesis of no relationship between gender and preference for learning English through online
classes and conclude that there is no statistically significant relationship between the two variables.
It is important to note that the study only investigated one aspect of student preferences, and further
research is needed to explore other factors that may affect students' preferences for online English
language learning. However, in conclusion, the lack of significant relationship implies that gender
does not influence students' preference for learning English through online classes.
Chi-square - Practice
A university language department wants to investigate whether there is a relationship
between students' language proficiency level (beginner, intermediate, or advanced)
and their preferred language learning activities (reading, writing, speaking, or listening).

Reading Writing Speaking Listening

Beginner 10 20 20 10

Intermediate 20 25 15 10

Advanced 25 15 20 10

Can we reject the null hypothesis of no difference at an .05 level? Write a report
 Chi-square - Practice
Row marginals
Reading Writing Speaking Listening
(Totals)

Beginner 10 20 20 10 60

Intermediate 20 25 15 10 70

Advanced 25 15 20 10 70

Column marginals
55 60 55 30 200
(Totals)
Chi-square - Practice
THANK YOU &
GOOD LUCK