University of Jeddah (UJ)

Faculty of Engineering

Engineering Statistics

Engineering major’s students and the impact of various factors on them

Ahmed Alghamdi 2240366

Sattam Alghnmi 2142891

Bandar Alzanbaqi 2241718

Section reference number: ENIE-232 Section: E62

Instructor: Dr. Haitham Bahaitham

 Question #1

1. Sample Size for Estimation

Method

Parameter Standard deviation

Distribution Normal

Standard deviation 0.51

Confidence level 95%

Confidence interval Two-sided

Results

Margin Sample

of Error Size

0.1 73
 Question #2

participants specialization % No. participants

Industrial Engineering 0.23 16.79

Chemical Engineering 0.30 21.90

Mechanical Engineering 0.22 16.06

Electrical Engineering 0.25 18.25

1. Industrial Engineering: 23% of 73 students ≈

0.23×73≈16.790.23×73≈16.79 (round up to 17 students)

2. Chemical Engineering: 30% of 73 students ≈ 0.30×73≈21.90.30×73≈21.9

(round up to 22 students)

3. Mechanical Engineering: 22% of 73 students ≈

0.22×73≈16.060.22×73≈16.06 (round up to 17 students)

4. Electrical Engineering: 25% of 73 students ≈

0.25×73≈18.250.25×73≈18.25 (round up to 18 students)

 Question # 3

Descriptive Statistics: Cumulative average (GPA), N0. of hours passed, Dist.

travel home to uni., No. of S.M(social media) apps on phone, minutes used S.M

weekly

Statistics

Variabl N N Mea SE StD Vari Coef Mini Q1

e * n Mea ev ance Var mum

Cumulat 4 0 4.38 0.09 0.6 0.430 14.98 3.000 4.0

ive 8 33 47 564 9 0 925

average(

GPA)

N0.of 4 0 87.5 3.49 24. 583.3 27.59 36.00 74.

hours 8 4 15 2 00

passed

Dist. trvl 4 0 33.8 2.66 18. 338.8 54.34 3.00 20.

hom to 8 8 41 8 00

uni.

No. of 4 0 7.56 0.99 6.8 47.27 90.92 3.000 4.0

S.M 8 3 2 76 3 00

apps on
phone

minutes 4 0 830 134 931 8665 112.0 90 300

used 8 21 9

S.M

weekly

Variable Media Q3 Maxi

n mum

Cumulati 4.6550 4.88 5.0000

ve 00

average(

GPA)

N0.of 76.50 106. 145.00

hours 75

passed

Dist. trvl 29.00 49.7 90.00

hom to 5

uni.

No. of 5.000 7.00 40.000

S.M apps 0

on phone

minutes 480 990 5400

used S.M

weekly
3. Regression Analysis: Cumulative average(GPA) versus N0.of hours

passed, Dist. trvl hom to uni., No. of S.M apps on phone, minutes used

S.M weekly

Regression Equation

Cumulative = 4.876 + 0.00017 N0.of hours passed

average(GPA) + 0.00251 Dist. trvl hom to uni.

- 0.02214 No. of S.M apps on phone

- 0.000511 minutes used S.M weekly

Coefficients

Term Coef SE Coef T-Value P-Value VIF

Constant 4.876 0.257 18.96 0.000

N0.of hours passed 0.00017 0.00233 0.07 0.944 1.06

Dist. trvl hom to uni. 0.00251 0.00300 0.84 0.408 1.02

No. of S.M apps on phone -0.02214 0.00849 -2.61 0.013 1.14

minutes used S.M weekly - 0.00006 -8.22 0.000 1.12

0.000511 2

Model Summary

S R-sq R-sq(adj) R-sq(pred)

0.375360 70.09% 67.30% 43.69%

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value

Regression 4 14.1946 3.54865 25.19 0.000

N0.of hours passed 1 0.0007 0.00071 0.01 0.944

Dist. trvl hom to uni. 1 0.0983 0.09826 0.70 0.408

No. of S.M apps on phone 1 0.9575 0.95748 6.80 0.013

minutes used S.M weekly 1 9.5191 9.51914 67.56 0.000

Error 43 6.0585 0.14089

Total 47 20.2531

Fits and Diagnostics for Unusual Observations

 Cumulative

Obs average(GPA) Fit Resid Std Resid

18 4.040 4.791 -0.751 -2.04 R

26 3.300 4.029 -0.729 -2.00 R

31 3.000 2.028 0.972 4.01 R X

37 3.500 3.531 -0.031 -0.12 X

38 3.100 4.075 -0.975 -2.65 R

R Large residual

X Unusual X
 Question # 5

Descriptive Statistics Summary

 Mean (4.38): The average GPA of the students is 4.38, indicating that, on

average, students perform quite well.

 Standard Error (0.0947): The standard error of the mean is 0.0947,

suggesting that the sample mean is a precise estimate of the population

mean.

 Median (4.655): The median GPA is 4.655, which is higher than the

mean, indicating a skewness in the distribution.

  Mode (5): The most frequently occurring GPA is 5, the highest possible

value.

 Standard Deviation (0.656): This value indicates the GPA values are

spread out around the mean. A standard deviation of 0.656 suggests

moderate variability.

 Sample Variance (0.431): This is the square of the standard deviation and

further indicates the spread of GPA values.

 Kurtosis (-0.256): Negative kurtosis indicates a slightly flatter

distribution than a normal distribution, with lighter tails.

 Skewness (-1.104): The negative skewness indicates that the GPA

distribution is left-skewed, meaning more students have higher GPAs.

 Range (2): The difference between the maximum and minimum GPA is 2.

 Minimum (3): The lowest GPA recorded is 3.

 Maximum (5): The highest GPA recorded is 5.

 Sum (210.4): The total sum of GPAs for all students.

 Count (48): The number of students in the sample.

Comments on Results

1. High Mean and Median: Both the mean (4.38) and median (4.655) GPAs

are relatively high, suggesting that students in this group generally

perform well academically.

2. Negative Skewness: The negative skewness (-1.104) indicates that the

GPA distribution is skewed towards higher values, which means that a

significant number of students have GPAs closer to the upper end of the

scale (5.0).

3. Moderate Variability: The standard deviation (0.656) shows that while

there is some variability in student GPAs, most students have GPAs

clustered around the mean.

4. High Mode: The mode of 5.0, being the highest GPA possible, shows that

this GPA value is quite common among students, reflecting a strong

academic performance.

5. Flat Distribution: The slight negative kurtosis (-0.256) suggests the

distribution is relatively flat compared to a normal distribution, indicating

a wider spread of GPA values but with lighter tails.

Question # 6
Student GPA and number of credit hours completed

The regression line (red line) represents the overall trend in the data using a linear

model. Here are the steps to interpret this:

1. Slope of the Regression Line:

1) If the slope of the regression line is positive, it indicates that, on

average, the GPA increases as the number of credit hours passed

increases.

2) If the slope is negative, it indicates that, on average, the GPA

decreases as the number of credit hours passed increases.

From the scatterplot, the regression line appears to have a slight upward slope,

suggesting a general increase in GPA as the number of credit hours increases.

However, the increase is not steep, indicating a relatively weak positive trend.
Student GPA and distance home to university

The regression line (red line) on the scatterplot represents the overall trend in the

data using a linear model:

 Slope of the Regression Line:

1) Positive Slope: If the slope of the regression line is positive, it

indicates that, on average, the GPA increases as the university

commuting distance increases.

2) Negative Slope: If the slope is negative, it indicates that, on

average, the GPA decreases as the university commuting distance

increases.

From the scatterplot, the regression line appears to have a slight upward slope,

suggesting a general decrease in GPA as the university commuting distance

increases. However, the increase is not steep, indicating a relatively weak positive

trend.
Student GPA and number of social media apps installed in student’s phone

The regression line (red line) on the scatterplot represents the overall trend in the

data using a linear model:

 Slope of the Regression Line:

1) Positive Slope: If the slope of the regression line is positive, it

indicates that, on average, the GPA increases as the number of

social media apps installed on the phone increases.

2) Negative Slope: If the slope is negative, it indicates that, on

average, the GPA decreases as the number of social media apps

installed on the phone increases.

From the scatterplot, the regression line has a negative slope, suggesting a general

decrease in GPA as the number of social media apps on the phone increases. This

indicates

negative

correlation between the number of social media apps and GPA.

Student GPA and weekly average time spent on social media apps

The regression line (red line) on the scatterplot represents the overall trend in the

data using a linear model:

 Slope of the Regression Line:

1) Positive Slope: If the slope of the regression line is positive, it

indicates that, on average, the GPA increases as the weekly average

time spent on social media apps.

2) Negative Slope: If the slope is negative, it indicates that, on

average, the GPA decreases as the weekly average time spent on

social media apps.

 From the scatterplot, the regression line has a negative slope,

suggesting a general decrease in GPA as the weekly average time

spent on social media apps is increased.

 . This indicates a negative correlation between the weekly average

time spent on social media and GPA.

Number of social media apps installed in student’s phone and average weekly

social media use

 The scatterplot shows a clear negative correlation between the number of

social media apps and the cumulative average GPA. As students install

more social media apps, their GPA tends to decrease.

 This trend suggests that high social media usage, which could be inferred

from having many social media apps, negatively impacts academic

performance.
 Question # 7

Method

Correlation type Pearson

Rows used 48

Correlations

Cumulative

average(GPA)

No. of S.M apps on phone -0.459

Pairwise Pearson Correlations

95% CI for

Sample 1 Sample 2 Correlation ρ P-Value

No. of S.M apps on Cumulative average -0.459 (-0.657, - 0.001

phone (GPA) 0.201)

1. GPA vs. Number of Social Media Apps Installed (r = -0.459):

 The Pearson correlation coefficient (r) of -0.459 indicates a

moderate negative correlation between the number of social media

apps installed on a student's phone and their GPA.

 This means that as the number of social media apps increases, the

GPA tends to decrease.

 The relationship is not extremely strong but suggests that social

media app usage has a noticeable impact on GPA.

7. Correlation: Cumulative average (GPA), minutes used S.M weekly

Method

Correlation type Pearson

Rows used 48

ρ: pairwise Pearson correlation

Correlations

Cumulative

average(GPA)

minutes used S.M weekly -0.803

Pairwise Pearson Correlations

Sample 1 Sample 2 Correlation 95% CI P-

for ρ Value

minutes used S.M Cumulative -0.803 (-0.885, - 0.000

 weekly average(GPA) 0.672)

2. GPA vs. Average Weekly Social Media Use (r = -0.803):

 The Pearson correlation coefficient (r) of -0.803 indicates a strong

negative correlation between the average weekly social media use

and GPA.

 This means that as the average time spent on social media per

week increases, the GPA significantly decreases.

 This strong negative correlation suggests that extensive time spent

on social media is strongly associated with lower academic

performance.

Question #8

1. GPA vs. Number of Social Media Apps Installed

 Correlation Coefficient (r): -0.459

 p-value: 0.001

2. GPA vs. Average Weekly Social Media Use

 Correlation Coefficient (r): -0.803

 p-value: <0.000

Interpretation of Results

GPA vs. Number of Social Media Apps Installed (r = -0.459)

 p-value = 0.004: Since the p-value is less than 0.05, we reject the

null hypothesis.

Conclusion: There is a statistically significant moderate negative

correlation between the number of social media apps installed and

GPA. This suggests that the number of social media apps is

meaningfully related to GPA, with more apps being associated

with lower GPA.

1. GPA vs. Average Weekly Social Media Use (r = -0.803)

 p-value < 0.001: Since the p-value is much less than 0.05, we

reject the null hypothesis.

 Conclusion: There is a statistically significant strong negative

correlation between average weekly social media use and GPA.

This suggests a very strong relationship where increased time spent

on social media is associated with significantly lower GPA.

Question # 9
9. normality test

9. one sample t test

Descriptive Statistics

SE 5% Lower Bound

N Mean StDev Mean for μ

48 7.563 6.876 0.992 9.228

μ: mean of No. of S.M apps on phone

Test

Null hypothesis H₀: μ =

Alternative H₁: μ >

hypothesis 3

T- P-

Value Value

4.60 0.000

A p-value of 0.000 indicates that the observed sample mean is significantly

different from the hypothesized mean (3) at the chosen significance level

(α=0.05). In other words, there is strong evidence to reject the null hypothesis that

the average number of social media apps installed on students' phones is 3.

Since the p-value is effectively zero (0.000), it means that the probability of

observing the sample mean or a more extreme value under the assumption that the

true population mean is 3 is extremely low. This suggests that the sample mean is

significantly greater than 3.

Therefore, based on the extremely low p-value, we reject the null hypothesis and

conclude that the average number of social media apps installed on students'

phones is statistically significantly greater than 3.

Question # 10
10. Variances: minutes used S.M weekly_1 versus specialization

Method

Null hypothesis All variances are equal

Alternative hypothesis At least one variance is different

Significance level α = 0.05

Tests

Test

Method Statistic P-Value

Multiple comparisons — 0.000

Levene 1.90 0.127

Samples are omitted from the tests if their standard deviations are 0 or missing.

10. One-way ANOVA: minutes used S.M weekly_1 versus specialization

Method

Null hypothesis All means are equal

Alternative hypothesis Not all means are equal

Significance level α = 0.05

Equal variances were assumed for the analysis.

Factor Information

Factor Levels Values

 specialization 23 3.00, 3.10, 3.14, 3.30, 3.50, 3.52, 3.60, 4.04, 4.50,

4.55, 4.58,

4.61, 4.64, 4.67, 4.70, 4.73, 4.80, 4.82, 4.83, 4.88,

4.89, 4.95,

5.00

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value

specialization 22 34518365 1569017 4.93 0.001

Error 17 5406919 318054

Total 39 39925284

Model Summary

S R-sq R-sq(adj) R-sq(pred)

563.963 86.46% 68.93% *

Means

specialization N Mean StDev 95% CI

3.00 3 300.0 60.0 (-387.0, 987.0)

3.10 1 180.0 * (-1009.9, 1369.9)

3.14 1 420.0 * (-769.9, 1609.9)

3.30 2 810 806 (-31, 1651)

3.50 2 1410 552 (569, 2251)

3.52 1 240.0 * (-949.9, 1429.9)

3.60 2 510 297 (-331, 1351)

 4.04 1 5400 * (4210, 6590)

4.50 2 195 148 (-646, 1036)

4.55 1 240.0 * (-949.9, 1429.9)

4.58 1 600.0 * (-589.9, 1789.9)

4.61 1 297.0 * (-892.9, 1486.9)

4.64 1 345.0 * (-844.9, 1534.9)

4.67 1 240.0 * (-949.9, 1429.9)

4.70 5 1151 874 (619, 1683)

4.73 1 3300 * (2110, 4490)

4.80 2 390.0 127.3 (-451.4, 1231.4)

4.82 1 855.0 * (-334.9, 2044.9)

4.83 1 90.00 * (-1099.86, 1279.86)

4.88 3 770 754 (83, 1457)

4.89 1 1680 * (490, 2870)

4.95 1 600.0 * (-589.9, 1789.9)

5.00 5 492.0 177.0 (-40.1, 1024.1)

Pooled StDev = 563.963

Tukey Pairwise Comparisons

Grouping Information Using the Tukey Method and 95% Confidence

specialization N Mean Grouping

4.04 1 5400 A

4.73 1 3300 A B

4.89 1 1680 B C
3.50 2 1410 B C

4.70 5 1151 B C

4.82 1 855.0 B C

3.30 2 810 B C

4.88 3 770 B C

4.95 1 600.0 B C

4.58 1 600.0 B C

3.60 2 510 B C

5.00 5 492.0 C

3.14 1 420.0 B C

4.80 2 390.0 B C

4.64 1 345.0 B C

3.00 3 300.0 C

4.61 1 297.0 B C

4.67 1 240.0 B C

4.55 1 240.0 B C

3.52 1 240.0 B C

4.50 2 195 C

3.10 1 180.0 B C

4.83 1 90.00 B C
 Means that do not share a letter are significantly different.

Question # 11

11. One-way ANOVA: Cumulative average (GPA) versus No. of S.M apps on

phone

Method

Null hypothesis All means are equal

 Alternative hypothesis Not all means are equal

Significance level α = 0.05

Equal variances were assumed for the analysis.

Factor Information

Factor Levels Values

No. of S.M apps on phone 14 3, 4, 5, 6, 7, 8, 9, 10, 15, 17, 18, 20, 26,

40

Analysis of Variance

Source DF Adj Adj F- P-

SS MS Value Value

No. of S.M apps on 13 12.84 0.9884 4.54 0.000

phone 9

Error 34 7.404 0.2178

Total 47 20.25

Model Summary

S R-sq R-sq(adj) R-sq(pred)

0.466646 63.44% 49.47% *

Means

No. of N Mean StDev 95% CI

S.M apps
on phone

3 7 4.757 0.334 (4.399, 5.116)

4 12 4.415 0.587 (4.141, 4.689)

5 6 4.557 0.570 (4.170, 4.944)

6 7 4.7757 0.2434 (4.4173, 5.1342)

7 5 4.6200 0.1829 (4.1959, 5.0441)

8 1 4.700 * (3.752, 5.648)

9 2 3.625 0.884 (2.954, 4.296)

10 1 3.100 * (2.152, 4.048)

15 2 3.150 0.212 (2.479, 3.821)

17 1 4.730 * (3.782, 5.678)

18 1 3.140 * (2.192, 4.088)

20 1 3.000 * (2.052, 3.948)

26 1 4.530 * (3.582, 5.478)

40 1 3.500 * (2.552, 4.448)

Pooled StDev = 0.466646

11. Regression Analysis: Cumulative average (GPA) versus N0. of hours

passed, Dist. travel home to uni., No. of S.M apps on phone, minutes

used S.M weekly

Regression Equation

Cumulative = 4.876 + 0.00017 N0.of hours passed

average(GPA) + 0.00251 Dist. trvl hom to uni.

- 0.02214 No. of S.M apps on phone

- 0.000511 minutes used S.M weekly

Coefficients

Term Coef SE Coef T-Value P-Value VIF

Constant 4.876 0.257 18.96 0.000

 N0.of hours passed 0.00017 0.00233 0.07 0.944 1.06

Dist. trvl hom to uni. 0.00251 0.00300 0.84 0.408 1.02

No. of S.M apps on -0.02214 0.00849 -2.61 0.013 1.14

phone

minutes used S.M -0.000511 0.000062 -8.22 0.000 1.12

weekly

Model Summary

S R-sq R-sq(adj) R-sq(pred)

0.375360 70.09% 67.30% 43.69%

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value

Regression 4 14.1946 3.54865 25.19 0.000

N0.of hours passed 1 0.0007 0.00071 0.01 0.944

Dist. trvl hom to uni. 1 0.0983 0.09826 0.70 0.408

No. of S.M apps on 1 0.9575 0.95748 6.80 0.013

phone

minutes used S.M 1 9.5191 9.51914 67.56 0.000

weekly

Error 43 6.0585 0.14089

Total 47 20.2531
Fits and Diagnostics for Unusual Observations

Obs Cumulative Fit Resid Std Resid

average(GPA)

18 4.040 4.791 -0.751 -2.04 R

26 3.300 4.029 -0.729 -2.00 R

31 3.000 2.028 0.972 4.01 R X

37 3.500 3.531 -0.031 -0.12 X

38 3.100 4.075 -0.975 -2.65 R

R Large residual

X Unusual X
 11. One-way ANOVA: Cumulative average (GPA) versus minutes

used S.M weekly

Method

Null hypothesis All means are equal

Alternative hypothesis Not all means are equal

Significance level α = 0.05

Equal variances were assumed for the analysis.

Factor Information

Factor Levels Values

minutes used S.M 27 90, 180, 236, 240, 297, 300, 345, 360, 420,

weekly 480, 510, 600,

660, 720, 780, 855, 900, 1020, 1200, 1380,

1620, 1680, 1800,

1980, 2160, 3300, 5400

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value

minutes used S.M weekly 26 18.804 0.72324 10.48 0.000

Error 21 1.449 0.06899

Total 47 20.253

Model Summary

S R-sq R-sq(adj) R-sq(pred)

0.262662 92.85% 83.99% *

Means

minutes N Mean StDev 95% CI

used S.M

weekly

90 2 5.000 0.000 (4.614, 5.386)

180 2 5.000 0.000 (4.614, 5.386)

236 1 5.000 * (4.454, 5.546)

240 5 4.728 0.402 (4.484, 4.972)

297 1 4.500 * (3.954, 5.046)

300 5 4.8480 0.0438 (4.6037, 5.0923)

345 1 4.640 * (4.094, 5.186)

360 1 4.900 * (4.354, 5.446)

420 2 4.7600 0.0849 (4.3738, 5.1462)

480 5 4.7640 0.0902 (4.5197, 5.0083)

510 1 4.460 * (3.914, 5.006)

600 3 4.5633 0.0416 (4.2480, 4.8787)

660 1 4.730 * (4.184, 5.276)

720 1 4.500 * (3.954, 5.046)

780 1 4.250 * (3.704, 4.796)

855 1 4.370 * (3.824, 4.916)

900 3 4.533 0.180 (4.218, 4.849)

1020 1 3.500 * (2.954, 4.046)

1200 2 4.050 0.750 (3.664, 4.436)

1380 2 3.350 0.354 (2.964, 3.736)

1620 1 3.140 * (2.594, 3.686)

1680 1 3.300 * (2.754, 3.846)

1800 1 3.500 * (2.954, 4.046)

1980 1 3.300 * (2.754, 3.846)

2160 1 3.000 * (2.454, 3.546)

3300 1 3.000 * (2.454, 3.546)

5400 1 3.000 * (2.454, 3.546)

Pooled StDev = 0.262662