INDEPENDENT SAMPLE T-

TEST
What is t-test?
▪Is used to compare data that we get in two groups
▪The t-statistic shows if there is a significant
difference (not expected difference) between the
two groups
▪For example: Is life satisfaction of students high
when they experience counseling?
Research Problem
We have 20 students; 10 of them experience
counseling while the other 10 students doesn’t
experience yet counseling.
- We measure their life satisfaction to know if
there’s a difference between the two groups
Hypotheses
H0 = There is no significant difference in the life satisfaction between the two groups.
(u1 = u2)
Ha = There is a significant difference in the life satisfaction between the two groups.
(u1 ≠ u2)
Data
Undergone Counselling Never experience counselling
7 5
8 4
10 7
8 6
7 7
6 5
4 3
7 4
8 4
9 5
Step 1: Compute the mean of the two groups
Undergone Counselling (X1) Never experience counselling (X2)
7 5
8 4
10 7
8 6
7 7
6 5
4 3
7 4
8 4
9 5
TOTAL = 74 TOTAL = 50
M = 74/10 = 7.4 M = 50/10 = 5
Step 2: Compute the variance of each group
X1 X1^2 X2 X2^2

7 49 5 25
8 64 4 16
10 100 7 49
8 64 6 36
7 49 7 49
6 36 5 25
4 16 3 9
7 49 4 16
8 64 4 16
9 81 5 25
TOTAL = 74 ΣX1^2=572 TOTAL = 50 ΣX2^2=266
M = 74/10 = 7.4 M = 50/10 = 5
Step 2: Compute the Variance of each
group
Step 3: Compute the Standard error of
difference between means
Step 4: Compute for T-statistics
Step 5: Know the critical value
Step 6: Compare the t-statistic (computed
value) to the critical value (from the table)
t = 3.58
Table t = 2.101
If the the t statistic (computed table ) is much
greater than the table t-value we will reject the null
hypothesis of H0
Conclusion: Therefore, there is significant difference
between the two groups
DEPENDENT/PAIRED
SAMPLE T-TEST
Hypothesis
H0: There is no significant difference in life satisfaction scores before and during
the COVID-19 pandemic
Two-tailed hypothesis
H1: There is significant difference in life satisfaction scores before and during
COVID -19 pandemic
One –tailed hypothesis
H1: The life satisfaction scores obtained before the COVID-19 pandemic will be
higher compared to the scores obtained during the outbreak
Result
This study is conducted to determine if there are changes in
life satisfaction scores before and during the COVID 19
pandemic. A paired samples t-test was utilized since the
researchers utilized a within subjects design. Results reveal
that the scores before COVID (M=109,4, s =8.08) was
significantly higher compared to the scores obtained during
the outbreak (M=91.7, s=5.52), t(14) = 6.76 , p < .001,
d=1.74. With these results, it can be argued that the COVID-
19 pandemic may have an effect on a persons well being or
happiness
ONE SAMPLE T-TEST
One sample t-test
• Parametric statistical tool that allows us to learn whether a sample
differs from the population from which it was drawn when the
population standard deviation is not known (Christopher,2017)
Formula
Sample Problem
NUMBER OF DAYS (x)
On average, the dogs in a local animal rescue
organization get adopted after 22 days (u=22 30
days) . Some digs get adopted in less than 22 35
days while some take more than 22 days to
20
get adopted. The Black Dog Syndrome refers
to the tendency for black dogs to get passed 41
over for adoption in favor of nonblack dogs. 58
Determine if there is a significant difference in
the number of days it takes for black dogs to 2
be adopted compared to other breeds 18
28
State the hypotheses
H0 : μ black dogs = μ all dogs
H1 : μ black dogs ≠ μ all dogs

α = 0.05 (two-tailed)
Locate the critical region
• α = 0.05 (two-tailed)
• Df = n -1
= 8-1
=7
• Critical value = 2.365
Make a decision
T = | 1.18|
Critical value = |2.365|

The number of days it takes for black dogs to get adopted (M=29, SD
16.71) does not significantly differ from that of other types of digs,
t=1.18, p =275.
ONE-WAY ANOVA
What is ANOVA?
• The ANOVA or F-test is used to compare 3 or more data in a
groups
• This shows if theres a significant difference in these groups
• For example : What is the ideal temperature to take an exam?
COLD, NORMAL (room temperature) or HOT?
State the hypotheses
H0 : μ 1 = μ 2 = :μ 3
H1 : At least one group is different from another

α = 0.05 (two-tailed)
DATA
COLD (15 C) NORMAL (25 C) HOT (35 C)
4 8 2
7 6 2
5 7 3
3 6 4
5 5 1
Step 1: Compute the mean of the groups

COLD (15 C) NORMAL (25 C) HOT (35 C)

4 8 2
7 6 2
5 7 3
3 6 4
5 5 1
M = 4.8 M=6.4 M=2.4
Step 2: Complete the table
k = number of
X1 (15 C) X1^2 X2 (25 C) X2 ^2 X3 (35 C) X3^2 treatment conditions
4 16 8 64 2 4 K n = number of scores
in each treatment
7 49 6 36 2 4 n N = total number of
scores
5 25 7 49 3 9 N T= total for each
3 9 6 36 4 16 G
treatment condition
G= sum of all scores in
5 25 5 25 1 1 ΣX^2 the study
SS=sum of squares
Total = ΣX1^2=1 T=32 ΣX2^2=2 T=12 ΣX3^2=3
24 24 10 4

M = 4.8 M=6.4 M=2.4

SS = 8.8 SS = 5.2 SS = 5.2

Step 3: Complete the ANOVA table
SOURCE SS Df MS F
Between 40.53 2
Treatments
Within 19.2 12
Treatments
Total 59.73 14
Step 3: Complete the ANOVA table
SOURCE SS Df MS F
Between 40.53 2
Treatments
Within 19.2 12
Treatments
Total 59.73 14
Step 3: Complete the ANOVA table
SOURCE SS Df MS F
Between 40.53 2 20.265 12.67
Treatments
Within 19.2 12 1.6
Treatments
Total 59.73 14
Step 4: Find the tabular value for F
Step 5. Compare the F-statistic (computed value) to
critical value
F=12.67
Table F = 3.88
If the F is greater than Table F we will reject the H0 or null
hypothesis
Conclusion: There is Significant difference in exam score among 3
groups of students in different temperature
Correlation
Pearson-r
• Correlation is a statistical technique that is used to
measure and describe the relationship between
two variables
The strength of relationship
Statistical analysis can show the strength of relationship between 2
variables

MODERATE CORRELATION WEAK CORRELATION

The strength of relationship
Correlation coefficients
range from -1.00 to
+1.00 coefficients
closer to 1 (- or +) are
stronger .
The Problem
• Does family income influence the average grade of the student?

• Null hypothesis: There is no relationship between family income

and grade.
• Alternative hypothesis: There is a relationship between family
income and grade.
Reporting the results
• Decision : Reject the Null Hypothesis
• Interpretation: There is a strong positive relationship (
r=.65) between family income and the student's average
grade
REGRESSION
What is regression?
This is an extension of correlation analysis
This is use to predict one variable
For example: How much is the fare of riding
a jeep if the travel is 5kms long?
Sample Results