1

Results

Name

Institution Affiliation
 2

t-Statistic Degrees p-Value Effect Size

of (Cohen's

Std. Freedom d)

N Mean Deviation
Sex Female 384 6.177 2.2096 -0.020 798 0.984 -0.001
Male 416 6.181 2.2028
Smoking No 395 6.104 2.1016 -0.946 798 0.344 -0.067
Yes 405 6.252 2.3011
Physical No 400 6.194 2.2041 0.186 798 0.853 0.013
activity

Yes 400 6.165 2.2080

Descriptive statistics
Glucose

Mean 6.179
0.07794
Standard Error 8
Median 5.9
Mode 6.4
Standard 2.20470
Deviation 2
Sample Variance 4.86071
Kurtosis -0.02089
0.81569
Skewness 6
Range 9
Minimum 3
Maximum 12
Sum 4943.2
Count 800
 3

Histogram

Frequency
120
110
100 106

90
80 84
81
70 74
71
65 Frequency
60 64 62
50
40
30
20 26
23 21
20 20 18
17 16
10 13 13
0 6
1 - 3 3.1 - 3.6 - 4.1 - 4.6 - 5.1 - 5.6 - 6.1 - 6.6 - 7.1 - 7.6 - 8.1 - 8.6 - 9.1 - 9.6 - 10.1 10.6 11.1 11.6
3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 - - 11 - - 12
10.5 11.5

SPSS Analysis observation

Association between the outcome variable (glucose level) and sex

 4

Group by Sex (0 = Female, 1= Male)

Female
50.01% 49.99%
Male

In the analysis, the first step was to see the difference in the glucose level between the

two groups (male and female). The sample size for females was 384 while the sample size for

males was 416. Each group represented its mean and standard deviation where the female group

had 6.177 mmol/L mean and 2.2096 mmol/L as standard deviation. On the other hand males had

a mean of 6.181 mmol/L and a standard deviation of 2.2028 mmol/L. In this test we has a t-

stastic value of – 0.020 that is associated with 798 degrees of freedom for the assumed variance

(792.510 for equal variance not assumed). The Independent T-test showed a p-value of 0.984

having an effect size of – 0.001.

Association between the outcome variable (glucose level) and Smoking

 5

Group Smoking (0 = non-smokers, 1=

smokers)

No
48.78% Yes
51.22%

The second step was to see the difference in the glucose level between the two groups

(Non-smokers and smokers). The sample size for non-smokers was 395 while the sample size for

smokers was 405. Each group represented its mean and standard deviation where the non-

smokers group had 6.104 mmol/L mean and 2.1016 mmol/L as standard deviation. On the other

hand, smokers had a mean of 6.254 mmol/L and a standard deviation of 2.3011 mmol/L. In this

test we have a t-statistic value of – 0.020 that is associated with 798 degrees of freedom for the

assumed variance (794.592 for equal variance not assumed). The Independent T-test showed a p-

value of 0.344 having an effect size of – 0.067.

Association between the outcome variable (glucose level) and Physical activity
 6

Group by Physical activity (0 = non-

physically active group, 1= physically
active group)

Female
Male
47.99%
52.01%

The last step was to see the difference in the glucose level between the two groups (those

having physical activity as yes and those do not do physical activity as no). The sample size for

both was 400 each. Each group represented its mean and standard deviation where the no group

had 6.194 mmol/L mean and 2.3011 mmol/L as standard deviation. On the other hand, yes had a

mean of 6.165 mmol/L and a standard deviation of 2.2028 mmol/L. In this test we have a t-

statistic value of – 0.186 that is associated with 798 degrees of freedom for the assumed variance

(797.997 for equal variance not assumed). The Independent T-test showed a p-value of 0.853

having an effect size of 0.013.

Interpret and summarize your findings

Association between the outcome variable (glucose level) and sex

In the first hypothesis, p-value (0.984) is higher than the standard alpha threshold (0.05). It

indicates that there is no statistically significant difference in the fasting blood glucose levels

between males and females according to the independent samples t-test. The two groups' mean

fasting blood glucose levels were nearly equal, with males having 6.181 mmol/L and females
 7

having 6.177 mmol/L. This illustrates that the means have almost equal values and represent the

true test result. From one’s observation there is a small difference in the mean glucose between

male and females. Although from the observed mean we can see that females have slightly

higher mean value than males.

There is sufficient information to support there is no statistical difference of the means between

the female and males. The sample's fasting blood glucose levels are not expressively affected by

sex, as suggested by the Cohen's d effect size (-0.001). These findings imply that sex by itself

does not considerably affect fasting blood glucose levels in this sample population. As a result,

other variables may be more important to consider when figuring out and managing this

population's diabetes risk.

Association between the outcome variable (glucose level) and Smoking

The results shown above contain the independent samples t-test for Smoking as a factor of fating

glucose level. As we can see, the test is greater than the standard alpha threshold of 0.05. That is

to say that there was no statistically significant difference in the fasting blood glucose levels

between smokers and non-smokers (from the table above we see the p-value = 0.344 that is

greater than 0.05). Smokers had to some extent higher mean fasting blood glucose levels (6.252

mmol/L) than non-smokers (6.104 mmol/L), but there was no statistically significant difference

between the two groups. The support is draw from the fact that the p-value is observed to be

higher.

There is sufficient information to support there is no statistical difference of the means between

the smokers and non-smokers. We can see that the sample's fasting blood glucose levels are not

significantly affected by the smokers' status. This follows the suggestion from the effect size

Cohen's d effect size (-0.067). These findings show that smoking status by itself does not
 8

substantially affect fasting blood glucose levels in the context of this group. As a result, other

variables may be more important to consider when figuring out and managing this population's

diabetes risk.

Association between the outcome variable (glucose level) and Physical activity

A p-value of 0.853 is higher than the standard alpha threshold of 0.05.It indicates that the

independent samples t-test did not find any statistically significant difference in fasting blood

glucose levels between physically active and inactive persons. Between the two groups, the mean

fasting blood glucose levels were quite similar: 6.165 mmol/L for the physically active group

and 6.194 mmol/L for the non-physically active group. From one’s observation there is a small

difference in the mean glucose between physically active group and non-physically active group.

Although from the observed mean we can see that non-physically active group have slightly

higher mean value than physically active group.

There is sufficient information to support there is no statistical difference of the means between

the non-physically active group and physically active group. The physical activity status of the

sample has no discernible effect on fasting blood glucose levels, as suggested by the effect size

(Cohen's d) of 0.013, which indicates a very tiny effect. These findings imply that physical

activity level by itself does not substantially affect fasting blood glucose levels in the context of

this group. As a result, other variables may be more important to consider when figuring out and

managing this population's diabetes risk.

Appendix

T-TEST GROUPS=Sex(0 1)
/MISSING=ANALYSIS
 9

/VARIABLES=Glucose
/ES DISPLAY(TRUE)
/CRITERIA=CI(.95).

T-Test

Notes
Output Created 21-MAY-2024 02:53:18
Comments
Input Data C:\Users\Prince\Downloads\
DATA.sav
Active Dataset DataSet1
Filter <none>
Weight <none>
Split File <none>
N of Rows in Working Data 800
File
Missing Value Handling Definition of Missing User defined missing values
are treated as missing.
Cases Used Statistics for each analysis
are based on the cases with
no missing or out-of-range
data for any variable in the
analysis.
Syntax T-TEST GROUPS=Sex(0 1)
/MISSING=ANALYSIS
/VARIABLES=Glucose
/ES DISPLAY(TRUE)
/CRITERIA=CI(.95).
Resources Processor Time 00:00:00.00
Elapsed Time 00:00:06.39

[DataSet1] C:\Users\Prince\Downloads\DATA.sav
 10

Group Statistics
Sex (0 = Female, 1 = Male) N Mean Std. Deviation
Outcome Variable Y: Female 384 6.177 2.2096
(Fasting) blood glucose Male 416 6.181 2.2028
levels (mmol/L)

Group Statistics
Sex (0 = Female, 1 = Male) Std. Error Mean
Outcome Variable Y: (Fasting) blood Female .1128
glucose levels (mmol/L) Male .1080

Independent Samples Test

t-test for
Levene's Test for Equality of Equality of
Variances Means

F Sig. t
Outcome Variable Y: Equal variances assumed .008 .930 -.020
(Fasting) blood glucose Equal variances not -.020
levels (mmol/L) assumed

Independent Samples Test

t-test for Equality of Means

df Sig. (2-tailed) Mean Difference

Outcome Variable Y: (Fasting) Equal variances assumed 798 .984 -.0032
blood glucose levels (mmol/L) Equal variances not assumed 792.510 .984 -.0032

Independent Samples Test

t-test for Equality of Means
95% Confidence Interval of the
Std. Error Difference
Difference Lower Upper
Outcome Variable Y: Equal variances assumed .1561 -.3096 .3033
(Fasting) blood glucose Equal variances not .1561 -.3097 .3033
levels (mmol/L) assumed
 11

Independent Samples Effect Sizes

95%
Confidence
Interval
a
Standardizer Point Estimate Lower
Outcome Variable Y: Cohen's d 2.2061 -.001 -.140
(Fasting) blood glucose Hedges' correction 2.2082 -.001 -.140
levels (mmol/L) Glass's delta 2.2028 -.001 -.140

Independent Samples Effect Sizes

95% Confidence
Intervala
Upper
Outcome Variable Y: (Fasting) blood glucose Cohen's d .137
levels (mmol/L) Hedges' correction .137
Glass's delta .137

a. The denominator used in estimating the effect sizes.

Cohen's d uses the pooled standard deviation.
Hedges' correction uses the pooled standard deviation, plus a correction factor.
Glass's delta uses the sample standard deviation of the control group.

T-TEST GROUPS=Smoking(0 1)
/MISSING=ANALYSIS
/VARIABLES=Glucose
/ES DISPLAY(TRUE)
/CRITERIA=CI(.95).

T-Test

Notes
Output Created 21-MAY-2024 03:02:31
Comments
 12

Input Data C:\Users\Prince\Downloads\

DATA.sav
Active Dataset DataSet1
Filter <none>
Weight <none>
Split File <none>
N of Rows in Working Data 800
File
Missing Value Handling Definition of Missing User defined missing values
are treated as missing.
Cases Used Statistics for each analysis
are based on the cases with
no missing or out-of-range
data for any variable in the
analysis.
Syntax T-TEST
GROUPS=Smoking(0 1)
/MISSING=ANALYSIS
/VARIABLES=Glucose
/ES DISPLAY(TRUE)
/CRITERIA=CI(.95).
Resources Processor Time 00:00:00.00
Elapsed Time 00:00:00.05

Group Statistics
Smoking (0 = no, 1 = yes) N Mean Std. Deviation
Outcome Variable Y: No 395 6.104 2.1016
(Fasting) blood glucose Yes 405 6.252 2.3011
levels (mmol/L)

Group Statistics
Smoking (0 = no, 1 = yes) Std. Error Mean
Outcome Variable Y: (Fasting) blood No .1057
glucose levels (mmol/L) Yes .1143

Independent Samples Test

 13

t-test for
Levene's Test for Equality of Equality of
Variances Means

F Sig. t
Outcome Variable Y: Equal variances assumed 2.925 .088 -.946
(Fasting) blood glucose Equal variances not -.947
levels (mmol/L) assumed

Independent Samples Test

t-test for Equality of Means

df Sig. (2-tailed) Mean Difference

Outcome Variable Y: (Fasting) Equal variances assumed 798 .344 -.1475
blood glucose levels (mmol/L) Equal variances not assumed 794.592 .344 -.1475

Independent Samples Test

t-test for Equality of Means
95% Confidence Interval of the
Std. Error Difference
Difference Lower Upper
Outcome Variable Y: Equal variances assumed .1559 -.4536 .1585
(Fasting) blood glucose Equal variances not .1557 -.4533 .1582
levels (mmol/L) assumed

Independent Samples Effect Sizes

95%
Confidence
Interval
a
Standardizer Point Estimate Lower
Outcome Variable Y: Cohen's d 2.2048 -.067 -.206
(Fasting) blood glucose Hedges' correction 2.2069 -.067 -.205
levels (mmol/L) Glass's delta 2.3011 -.064 -.203

Independent Samples Effect Sizes

95% Confidence
Intervala
Upper
Outcome Variable Y: (Fasting) blood glucose Cohen's d .072
levels (mmol/L) Hedges' correction .072
 14

Glass's delta .075

a. The denominator used in estimating the effect sizes.

Cohen's d uses the pooled standard deviation.
Hedges' correction uses the pooled standard deviation, plus a correction factor.
Glass's delta uses the sample standard deviation of the control group.

T-TEST GROUPS=Physical(0 1)
/MISSING=ANALYSIS
/VARIABLES=Glucose
/ES DISPLAY(TRUE)
/CRITERIA=CI(.95).

T-Test

Notes
Output Created 21-MAY-2024 03:04:20
Comments
Input Data C:\Users\Prince\Downloads\
DATA.sav
Active Dataset DataSet1
Filter <none>
Weight <none>
Split File <none>
N of Rows in Working Data 800
File
Missing Value Handling Definition of Missing User defined missing values
are treated as missing.
Cases Used Statistics for each analysis
are based on the cases with
no missing or out-of-range
data for any variable in the
analysis.
 15

Syntax T-TEST
GROUPS=Physical(0 1)
/MISSING=ANALYSIS
/VARIABLES=Glucose
/ES DISPLAY(TRUE)
/CRITERIA=CI(.95).
Resources Processor Time 00:00:00.00
Elapsed Time 00:00:00.07

Group Statistics
Physical activity (0 = no, 1 =
yes) N Mean Std. Deviation
Outcome Variable Y: No 400 6.194 2.2041
(Fasting) blood glucose Yes 400 6.165 2.2080
levels (mmol/L)

Group Statistics
Physical activity (0 = no, 1 = yes) Std. Error Mean
Outcome Variable Y: (Fasting) blood No .1102
glucose levels (mmol/L) Yes .1104

Independent Samples Test

t-test for
Levene's Test for Equality of Equality of
Variances Means

F Sig. t
Outcome Variable Y: Equal variances assumed .155 .694 .186
(Fasting) blood glucose Equal variances not .186
levels (mmol/L) assumed

Independent Samples Test

t-test for Equality of Means

df Sig. (2-tailed) Mean Difference

Outcome Variable Y: (Fasting) Equal variances assumed 798 .853 .0290
blood glucose levels (mmol/L) Equal variances not assumed 797.997 .853 .0290
 16

Independent Samples Test

t-test for Equality of Means
95% Confidence Interval of the
Std. Error Difference
Difference Lower Upper
Outcome Variable Y: Equal variances assumed .1560 -.2772 .3352
(Fasting) blood glucose Equal variances not .1560 -.2772 .3352
levels (mmol/L) assumed

Independent Samples Effect Sizes

95%
Confidence
Interval
a
Standardizer Point Estimate Lower
Outcome Variable Y: Cohen's d 2.2060 .013 -.125
(Fasting) blood glucose Hedges' correction 2.2081 .013 -.125
levels (mmol/L) Glass's delta 2.2080 .013 -.125

Independent Samples Effect Sizes

95% Confidence
Intervala
Upper
Outcome Variable Y: (Fasting) blood glucose Cohen's d .152
levels (mmol/L) Hedges' correction .152
Glass's delta .152

a. The denominator used in estimating the effect sizes.

Cohen's d uses the pooled standard deviation.
Hedges' correction uses the pooled standard deviation, plus a correction factor.
Glass's delta uses the sample standard deviation of the control group.