Ali Hussein – Research Technique – 2023/2024.

1: Descriptive Analysis of Sex (Nominal):

Statistics
Sex1
N Valid 20
Missing 0
Mode 2.00

Sex1
Cumulative
Frequency Percent Valid Percent Percent
Valid 1.00 6 30.0 30.0 30.0
2.00 14 70.0 70.0 100.0
Total 20 100.0 100.0
Subject: Descriptive Analysis of Sex of Students
Variable: Sex
Variable Type: Qualitative (Nominal)
N = Sample = 20 Students
Number of Females: 6
Number of Males: 14
Mode: Most Frequent Value of Student Sex (Highest Frequency) = M = 14
2: Descriptive Analysis of Satisfaction Level (Ordinal):

Statistics
Level1
N Valid 20
Missing 0
Median 2.00
Mode 1

Level1
Cumulative
Frequency Percent Valid Percent Percent
Valid 1 8 40.0 40.0 40.0
2 7 35.0 35.0 75.0
3 5 25.0 25.0 100.0
Total 20 100.0 100.0
N = Sample = 20 Student
Subject: Descriptive Analysis of Satisfaction Level of Students
Variable: Level
Variable Type: Qualitative (Ordinal)
Missing = Missing Value of Student = 0
Median: Middle Value of Set of Observation (Satisfaction Level) = 2 Medium
50% of the sample have a Satisfaction Level: Low, Medium
50% of the sample have s Satisfaction Level: Medium, High
Mode: Most Frequent Value of Student Sex (Highest Frequency) = M = 1 (Low)
3: Descriptive Analysis of Age (Scale):

Statistics
Age
N Valid 20
Missing 0
Mean 20.90
Std. Error of Mean .528
Median 21.00
Mode 20a
Std. Deviation 2.360
Variance 5.568
Skewness .026
Std. Error of Skewness .512
Kurtosis -.170
Std. Error of Kurtosis .992
Range 9
Minimum 16
Maximum 25
Sum 418
Percentiles 25 19.25
50 21.00
75 22.75
a. Multiple modes exist. The smallest
value is shown
N = Sample = 20 Student

Missing: Missing Value of Student = 0

Variable = Age

Variable Type: Quantitative (Scale or Numeric)

Mode: Most Frequent Value of Age (Highest Frequency) = 20

Median: Middle Value of Set of Observations (Age) = 21 years

50% of the sample have an age 21 and below

50% of the sample have an age 21 and above

Mean: Average of Student Age = 20.90 Years

Std Deviation: Measure of Dispersion Around the Mean = 2.36

Variance: 1/N * Sum (Xi – Mean)2: The Squared Spread from the Mean: Std Deviation2 = 5.568

Range: Maximum Value – Minimum Value = 25 – 16 = 9

Maximum Value: Max Age of Student = 25

Minimum Value: Minimum Age of Student = 16

Skewness = 0.026 = Asymmetric Coeff > 0

 Distribution to the Right

// Asymmetric > 0 = Right
// Asymmetric < 0 = Left

Kurtosis = -0.170 < 0 ➔ Flat

// if (Kurtosis < 0) ➔ Flat

// if (Kurtosis > 0) ➔ Peaky

i. Grouping the Satisfaction Level according the Sex and then Comment of the Results.

Analyze > Descriptive Statistics > Crosstabs > Put the Row Variable and the Column Variable > OK

Gender * Level of Satisfaction Crosstabulation

Count
Level of Satisfaction
High Low Medium Total
Gender F 0 5 1 6
M 5 3 6 14
Total 5 8 7 20

Females are distributed between Low and Medium Satisfaction Level

Most of the Males are distributed between High and Medium Satisfaction Level
➔ Males show better result toward weight loss than Females
4.One-Sample T-Test:

One Sample T-Tests can be used to determine if the mean of a sample is different from a particular value.

Question: Test if the Mean Value of the Age = 25:

Subject: Determine if the mean value of the Student’s Age = 25 years

N = Sample = 20 Student

Variable: Age

Variable Type: Quantitative (numeric)

Test: One Sample T-Test

Hypothesis:

 H0: Mean (Average) = 25 years ➔ No Statistical Significance

 H1: Mean (Average)! = 25 years ➔ It exists Statistical Significance

Interpretation:

One-Sample Statistics
N Mean Std. Deviation Std. Error Mean
Age 20 20.90 2.360 .528

One-Sample Test
Test Value = 25
95% Confidence Interval of the
Difference
t df Sig. (2-tailed) Mean Difference Lower Upper
Age -7.770 19 .000 -4.100 -5.20 -3.00

Result: Mean (Average) = 20.90! = 25

Confidence Interval = 95%

100% – 95% = 5% = 0.05 //dayman barren bl 0.05

Sig = 0.000 < 0.05 ➔ We Accept H1 ➔ It Exist a Statistical Significance

If (Sig < 0.05) ➔ Reject H0 (Accept H1)

If (Sig > 0.05) ➔ Fail to Reject H0 (Accept H0)

5. Binomial Test:

The Binomial Test is useful for determining if the proportion of people in one of two categories

(ex. YES/ NO, Male/ Female) is different from a specified amount.

Question: Test if The Proportion of Female Students = The Proportion of Male Students:

Subject: We should Test if the Proportion of Female Students = Proportion of Male Students

N = Sample = 20 students

Variable: Sex

Variable Type: Qualitative (Nominal)

Test: Binomial Test

Hypothesis:

 H0: Proportion (Male) = Proportion (Female) ➔ No Statistical Significance

 H1: Proportion (Male)! = Proportion (Female) ➔ It exist Statistical Significance

Interpretation:

Binomial Test
Exact Sig. (2-
Category N Observed Prop. Test Prop. tailed)
Sex1 Group 1 1 6 .30 .50 .115
Group 2 2 14 .70
Total 20 1.00

Group 1: Proportion of Females = 6 = 30%

Group 2: Proportion of Males = 14 = 70%

Sig = 0.115 > 0.05 ➔ We Accept H0

If (Sig < 0.05) ➔ Reject H0 (Accept H1)

If (Sig > 0.05) ➔ Fail to Reject H0 (Accept H0)

Result: Proportion (Male) = Proportion (Female) ➔ No Statistical Significance

6. Independent Sample T-Test:

The Independent Sample T-Test can be used to see if two means are different from each other when the
two samples that the means are based on were taken from different individuals who have not been
matched.
2 Variables: 1. One Variable Quantitative Dependent

2. One Variable Qualitative Independent contains 2 Groups

Question: Prove that WeightBefore is dependent (affected by) Gender

Subject: We should Prove that WeightBefore is dependent (affected by) Gender

Variables: WeightBefore and Gender

Variable Types: 1. WeightBefore (Quantitative and Dependent)

2. Gender (Qualitative and Independent, Contains 2 Groups (Male, Female))

Test: Independent Sample T-Test

Hypothesis:

 H0: Mean (WeightBefore of Females) = Mean (WeightBefore of Males) ➔ No Statistical

Significance
 H1: Mean (WeightBefore of Females)! = Mean (WeightBefore of Males) ➔ It exist Stat.
Significance

Interpretation:

Group Statistics

Gender N Mean Std. Deviation Std. Error Mean

Weight Before Exercises F 6 62.50 5.244 2.141
M 14 74.86 5.021 1.342

Mean (WeightBefore of Females) = 62.50 kg

Mean (WightBefore of Males) = 74.86 kg

  Mean (WeightBefore of Females)! = Mean (WeightBefore of Males)

Std. Deviation (WeightBefore of Females) = 5.244 ➔ It exists a certain dispersion around the mean

Std. Deviation (WeightBefore of Males) = 5.021 ➔ It exists a certain dispersion around the mean

Sig. = 0.890 > 0.05 ➔ We Read the Sig 2-Tailed of the first line

Sig. (2-tailed) = 0.00 < 0.05 ➔ We Accept H0

Result: Mean (WeightBefore of Females)! = Mean (WeightBefore of Males) ➔ It exist Stat. Significance

Significance ➔ The Gender has an effect on WeightBefore

7. One Way ANOVA Test:

2 Variables:

 1 Variable (Quantitative – Dependent)

 1 Variable (Qualitative – Independent)
 The Qualitative Variable Should Contains 3 Groups or more

Question: Prove that Weight After is Dependent (Affected by) Level of Satisfaction.

Subject: We should Prove that WeightAfter is Dependent (affected by) Level of Satisfaction

Variables: WeightAfter and Level

Variables Types: WeightAfter (Quantitative and Dependent)

Level (Qualitative and Independent)

Test: One Way ANOVA

Hypothesis:

HO: Mean (WeightAfter of Low) = Mean (WeightAfter of Medium) = Mean (WeightAfter of High)

➔No Statistical Significance

H1: At Least 2 Groups Have Different Means ➔ It exist Statistical Significance

Interpretation:

Test of Homogeneity of Variances

Levene Statistic df1 df2 Sig.
Weight After Exercises Based on Mean 2.442 2 17 .117
Based on Median 1.694 2 17 .213
Based on Median and with 1.694 2 16.987 .213
adjusted df
Based on trimmed mean 2.449 2 17 .116

ANOVA
Weight After Exercises
Sum of Squares df Mean Square F Sig.
Between Groups 84.525 2 42.263 .872 .436
Within Groups 823.675 17 48.451
 Total 908.200 19

Multiple Comparisons
Dependent Variable: Weight After Exercises

Mean Difference 95% Confidence Interval

(I) Level1 (J) Level1 (I-J) Std. Error Sig. Lower Bound Upper Bound
Tukey HSD 1 2 -.875 3.603 .968 -10.12 8.37
3 -5.075 3.968 .426 -15.25 5.10
2 1 .875 3.603 .968 -8.37 10.12
3 -4.200 4.076 .568 -14.66 6.26
3 1 5.075 3.968 .426 -5.10 15.25
2 4.200 4.076 .568 -6.26 14.66
Games-Howell 1 2 -.875 3.766 .971 -10.97 9.22
3 -5.075 3.798 .406 -15.36 5.21
2 1 .875 3.766 .971 -9.22 10.97
3 -4.200 2.904 .358 -12.23 3.83
3 1 5.075 3.798 .406 -5.21 15.36
2 4.200 2.904 .358 -3.83 12.23

Weight After Exercises

Subset for alpha
= 0.05
Level1 N 1
Tukey HSDa,b 1 8 67.13
2 7 68.00
3 5 72.20
Sig. .411

Table 1: Homogeneity

Sig = 0.117 > 0.05 ➔ The Groups are Homogenies

Table 2: ANOVA

Sig = 0.436 > 0.05 ➔ We Read the Result of Tukey in Post-Hoc Table

// if Sig > 0.05 ➔ Tukey in Post-Hoc

If Sig < 0.05 ➔ Games-Howell in Post-Hoc Table

Table 3: Post-Hoc
1*2 (Low * Medium): Sig = 0.968 > 0.05 ➔ We Accept H0
1*3 (Low * High): Sig = 0.426 > 0.05 ➔ We Accept H0
2*3(Medium * High): Sig = 0.568 > 0.05 ➔ We Accept H0

Result:
HO: Mean (WeightAfter of Low) = Mean (WeightAfter of Medium) = Mean (WeightAfter of High)
➔No Statistical Significance
 The WeightAfter is not Affected by the Level of Satisfaction
8. Paired Sample T-Test

2 Variables: Quantitative and Dependent

Question: Prove that the Exercises are an Effect on Weight

Subject: We should Prove that the Exercises are an Effect on Weight

Variable: WeightBefore and WeightAfter

Variable Types: Quantitative (Numeric)

Test: Paired Sample T-Test

Hypothesis:

H0: Mean (WeightBefore) = Mean (WeightAfter) ➔ No Statistical Significance

H1: Mean (WeightBefore)! = Mean (WeightAfter) ➔ It exist Statistical Significance

Interpretation:

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean
Pair 1 Weight Before Exercises 71.15 20 7.631 1.706
Weight After Exercises 68.70 20 6.914 1.546

Paired Samples Correlations

N Correlation Sig.
Pair 1 Weight Before Exercises & 20 .962 .000
Weight After Exercises
Mean (WeightBefore) = 71.15
Mean (WeightAfter) = 68.70
Std. Deviation = 7.631
Std. Deviation = 6.914
 There is a certain dispersion around the mean
Correlation = R = 0.962 ~ 1 ➔ Strong Relation Between WeightBefore and WeightAfter
Sig. = 0.000 < 0.05 ➔ We Accept H1
Result: H1: Mean (WeightBefore)! = Mean (WeightAfter) ➔ There is a Statistical Significance
9. Chi-Square Test:
2 Variables: Qualitative
Question: Test if Gender and Level of Satisfaction are Dependent
Subject: We should Test if Gender and Level of Satisfaction are Dependent
Variables: Gender and Level
Variable Types: Gender (Qualitative – Independent)
Level (Qualitative – Dependent)
Test: Chi-Square Test
Hypothesis:
H0: Variables are Independent ➔ No Statistical Significance
H1: Variables are Dependent ➔ It exists Statistical Significance
Interpretation:

Chi-Square Tests
Asymptotic
Significance (2-
Value df sided)
Pearson Chi-Square 6.990a 2 .030
Likelihood Ratio 8.108 2 .017
N of Valid Cases 20
a. 5 cells (83.3%) have expected count less than 5. The minimum
expected count is 1.50.

Sig = 0.030 < 0.05 ➔ We Accept H1

Result: H1: Variables are Dependent ➔ It exists Statistical Significance

10. Correlation Test:
2 Variables: Quantitative

Question: Prove that exists a Relation Between Age and WeightBefore

Subject: We should Prove that if exists a Relation Between Age and WeightBefore
Variables: Age and WeightBefore
Variable Types: Age (Quantitative – Independent)
WeightBefore (Quantitative – Dependent)
Test: Correlation Test
Hypothesis:
H0: There is a Relation Between Age and WeightBefore ➔ No Statistical Significance
H1: There is no Relation Between Age and WeightBefore ➔ It exist Stat. Significance
Interpretation:

Symmetric Measures
Asymptotic Approximate
Value Standard Errora Approximate Tb Significance
Interval by Interval Pearson's R .468 .219 2.250 .037c
Ordinal by Ordinal Spearman Correlation .429 .213 2.017 .059c
N of Valid Cases 20
a. Not assuming the null hypothesis.
b. Using the asymptotic standard error assuming the null hypothesis.
c. Based on normal approximation.

Pearson Correlation R = 0.37 < 0.3 ➔ Weak Correlation

// r < = 0.3 ➔ Weak Correlation

// 0.3 < r < 0.6 ➔ Middle Correlation

// r >= 0.6 ➔ Strong Correlation

Result: H1: No Relation Between Age and WeightBefore

11. Linear Regression Test:
2 Variables: Quantitative
It might be a Qualitative Variable (contains 5 groups or more)

Measuring the Relation Between Them.

Equation: y = ax + b
Question: Determine if it exist a relation between Weight Before and Age
Subject: We should determine if it exist a relation between Weight Before and Age
Variables: WeightBefore and Age
Variable Types: WeightBefore: Quantitative and Dependent
Age: Quantitative and Independent
Test: Linear Regression
Hypothesis:
H0: a = 0 ➔ No relation between the 2 Variables
H1: a != 0 ➔ It exist a relation between the 2 Variables
Interpretation:

Descriptive Statistics
Mean Std. Deviation N
Weight Before Exercises 71.15 7.631 20
Age 20.90 2.360 20

Correlations
Weight Before
Exercises Age
Pearson Correlation Weight Before Exercises 1.000 .468
Age .468 1.000
Sig. (1-tailed) Weight Before Exercises . .019
Age .019 .
N Weight Before Exercises 20 20
Age 20 20

Variables Entered/Removeda
Variables Variables
Model Entered Removed Method
1 Ageb . Enter
a. Dependent Variable: Weight Before Exercises
b. All requested variables entered.

Model Summaryb
Adjusted R Std. Error of the
Model R R Square Square Estimate
1 .468a .219 .176 6.927
a. Predictors: (Constant), Age
b. Dependent Variable: Weight Before Exercises

ANOVAa
Model Sum of Squares df Mean Square F Sig.
1 Regression 242.874 1 242.874 5.062 .037b
Residual 863.676 18 47.982
Total 1106.550 19
a. Dependent Variable: Weight Before Exercises
b. Predictors: (Constant), Age

Coefficientsa
Standardized
Unstandardized Coefficients Coefficients
Model B Std. Error Beta t Sig.
1 (Constant) 39.484 14.160 2.788 .012
Age 1.515 .673 .468 2.250 .037
a. Dependent Variable: Weight Before Exercises
 Residuals Statisticsa
Minimum Maximum Mean Std. Deviation N
Predicted Value 63.73 77.36 71.15 3.575 20
Residual -15.847 9.729 .000 6.742 20
Std. Predicted Value -2.076 1.737 .000 1.000 20
Std. Residual -2.288 1.404 .000 .973 20
a. Dependent Variable: Weight Before Exercises

Table 1: Descriptive Statistics

Mean of WeightBefore = 71.15kg
Mean of Age = 20 years
Table 2: Correlations
Pearson Correlation = 0.468
0.3 < 0.468 < 0.6 ➔ Middle Correlation
// r < = 0.3 ➔ Weak Correlation

// 0.3 < r < 0.6 ➔ Middle Correlation

// r >= 0.6 ➔ Strong Correlation

Table 3: Model Summary

R square = 0.219 = 21.9% of student have a dependency between WeightBefore and Age, then
other depend on other factors.
Table 4: ANOVA
Sig. = 0.37 < 0.05 ➔ We Accept H1 ➔ It exist a relation between the 2 Variables
Table 5: Coefficient
Y = ax + b
WeightBefore = 1.515 x + 39.484
A = 0.515 != 0 ➔ We have a relation between WeightBefore and Age