Abstract

[Draw your reader in with an engaging abstract. It is typically a short summary of the document.
When you’re ready to add your content, just click here and start typing.]

Microsoft
[Email address]
Part (a)
The mean annual passing rate for school of education is given by;
∑𝑛
𝑖=1 𝑥𝑖
𝑥̅ = where
𝑛1

𝑥̅ is mean annual passing rate of students in the school of Education,

∑ is the summation sign,
𝑛1 is the sample size (the number of years sampled in the school of Education) and
𝑥𝑖 denotes each observed data value.

19 + 8 +(−12) + (−3) + 15 + 18 + 20 + (−10) + (−6) + 16

Therefore, 𝑥̅ = 10
65
𝑥̅ = 10

𝑥̅ = 6.5%

The mean annual passing rate for the school of humanities is given by;
∑𝑛
𝑖=1 𝑦𝑖
𝑦̅= where
𝑛2

𝑦̅ is the mean annual passing rate of students in the school of Humanities,

∑ is the summation sign,
𝑛2 is the sample size (the number of years sampled from the school of Humanities), and
𝑦𝑖 denotes each observed data value.

8+3+(−9)+2+4+5+10+12+8+11
And so, 𝑦̅ = 10
54
𝑦̅ = 10

𝑦̅ = 5.4%

Page 1
Comment:
The mean annual passing rate of students from school of education is 1.1% more than that of the
students from the school of humanities. Therefore, it can be stated that the students from the
school of education have been performing better than the students from the school of humanities
in psychology final examinations during the past ten years.

Part (b)
The standard deviation for the annual passing rate in the school of education is given by;

∑(𝑥𝑖 −𝑥̅ )2
𝑠1 =√ 𝑛1 −1

But, 𝑥̅ = 6.5%, 𝑛1 =10

Table for the computation of the mean deviation squared.

year Passing Mean Mean

rate 𝑥𝑖 (%) deviation deviation
(𝑥𝑖 − 𝑥̅ ) squared
(𝑥𝑖 − 𝑥̅ )2
2007 19 12.5 156.25
2008 8 1.5 2.25
2009 -12 -18.5 342.25
2010 -3 -9.5 90.25
2011 15 8.5 72.25
2012 18 11.5 132.25
2013 20 13.5 182.25
2014 -10 -16.5 272.25
2015 -6 -12.5 156.25
2016 16 9.5 90.25
total 65 0 1496.5

Thus ∑(𝑥𝑖 − 𝑥̅ )2 = 1496.5 and so

∑(𝑥𝑖 −𝑥̅ )2 1496.5

𝑠1 =√ = √ 10−1
𝑛1 −1

𝑠1 = √166.2777778
𝑠1 = 12.89487409 = 12.9%
Page 2
Calculation of Standard deviation for the annual passing rate of students in the school of
Humanities.

∑(𝑦𝑖 −𝑦̅)2
𝑠2 =√ 𝑛2 −1

Given 𝑦̅= 5.4%, 𝑛2 =10

The table below shows the computation of the mean deviation squared.
year 𝑦𝑖 (%) 𝑦𝑖 − 𝑦̅ (𝑦𝑖 − 𝑦̅)2
2007 8 2.6 6.76
2008 3 -2.4 5.76
2009 -9 -14.4 207.36
2010 2 -3.4 11.56
2011 4 -1.4 1.96
2012 5 -0.4 0.16
2013 10 4.6 21.16
2014 12 6.6 43.56
2015 8 2.6 6.76
2016 11 5.6 31.36
totals 54 0 336.4

Thus ∑(𝑦𝑖 − 𝑦̅)2= 336.4 and so

∑(𝑦𝑖 −𝑦̅)2 336.4

𝑠2 =√ =√10−1
𝑛2 −1

𝑠2 = √37.7777778
𝑠2 = 6.11373681 = 6.11%

Comment:
Since the value of the standard deviation for the students from the school of education seems to
be higher than that of students from the school of humanities, it tells us that the dataset from the
school of education are much more spread out or dispersed about the mean than the data values
obtained from the school of humanities which seem to closely cluster about the mean value.

Page 3
Part (c):
Computation of a 95% confidence interval for the mean difference between the two schools.
Taking the sample size in Education to be 𝑛1 =10 and
Sample size in Humanities 𝑛2 =10,
Education sample mean 𝑥̅ = 6.5%,
Humanities sample mean 𝑦̅= 5.4%,
Education standard deviation 𝑠1 = 12.89487409,
Humanities standard deviation 𝑠2 = 6.11373681,
The significance level ∝ = 5%= 0.05.
For this confidence interval, the student’s t-distribution will be used because the population
variances 𝜎1 2 and 𝜎2 2 are unknown, then 𝑛1 and 𝑛2 are both less than 30. Assuming that the
populations have equal variances, the population variance 𝜎 2 will be estimated by pooled
variance 𝑠𝑝2 .
𝑋̅ −𝜇
The statistic T= 𝑆 ~𝑡𝑛−1 with n-1 degrees of freedom follows a student’s t-distribution.
√𝑛

Therefore, the confidence interval for the mean difference is given by

1 1
(𝑥̅ − 𝑦̅) ± 𝑡∝ , υ √𝑠𝑝2 ( 𝑛 + )
2 1 𝑛2

(𝑛1 −1)𝑠2 1 + (𝑛2 −1)𝑠2 2

Where 𝑠𝑝2 is pooled variance given by; 𝑠𝑝2 = 𝑛1 +𝑛2 −2

And υ = 𝑛1 + 𝑛2 − 2
(10−1)×(12.89487409)2 +(10−1)×(6.11373681)2
Therefore 𝑠𝑝2 = 10+10−2

𝑠𝑝2 = 101.8277778
And υ= 18
Hence the confidence interval is;

1 1
(6.5 − 5.4) ± 𝑡0.025 , 18 √101.8277778( 10 + )
10

1
= 1.1 ± 2.101 × √101.8277778 × 5

= (-8.3814, 10.5814)
Page 4
Hence with 95% confidence, it can be said that the difference in mean annual passing rates of
students between the two schools will fall within the interval (-8.3814, 10.5814).
QUESTION (d)
Hypothesis testing, two sample t –test
Given that;
Sample size in Education 𝑛1 = 10
Sample size in Humanities 𝑛2 = 10
Education sample mean 𝑥̅ = 6.5%
Humanities sample mean ̅𝑦= 5.4%
We will carry out this test step-by-step as outlined below and we will use 𝜇1 and 𝜇2 to denote
population means for school of Education and school of Humanities respectively.
1.0 Testing problem
In this context;
Null Hypothesis states that the two population means are not significantly different. That is;
𝐻0 : 𝜇1 = 𝜇2
Alternative Hypothesis states that the two population means are significantly different. That is;
𝐻𝐴 : 𝜇1 ≠ 𝜇2
1.2 Test statistic
𝑋̅ −𝑌̅
T= 1 1
√𝑆𝑃2 (𝑛 +𝑛 )
1 2

1.3 Critical region

Reject 𝐻𝑜 if |𝑡 ∗ | > 𝑡∝ , υ
2

That is if |𝑡 ∗ | > 𝑡0.025 , 18=2.101

1.4 Computation of the observed value 𝑡 ∗
6.5−5.4
𝑡 ∗= 1 1
√101.8277778( + )
10 10

𝑡 ∗ = 0.24374996
∴ 𝑡 ∗ = 0.2437
Page 5
That is |𝑡 ∗ |=|0.2437|=0.2437< 2.101
Fail to reject 𝐻𝑜

However, to obtain enough evidence, the p-value will be calculated. This is the probability
computed. Assuming that the null hypothesis is valid, the test statistic will take a value at least as
extreme as the observed one.
If the p-value is found to be less than the set level of significance, then there would be enough
evidence so that the null hypothesis should be rejected, but if the p-value is greater than the set
level of significance then there would be no evidence for the null hypothesis to be rejected.
Thus p-value = p (𝑇18 > |𝑡 ∗ |)
= p (𝑇18 > 0.2437)
The table for probability values shows that the p-value > 0.25 > 0.05
Therefore, the p-value is greater than the set level of significance i.e. p-value > 0.05, this means
that there is no enough evidence to reject the null hypothesis.

Conclusion
Fail to reject
Since the null hypothesis failed to reject at 𝐻0 at ∝ = 0.05, it means that the mean annual passing
rates of students for both schools are not significantly different.

Page 6
 REFERENCE
1. Hoel G.P. (1962) Introduction to Mathematical Statistics, 3𝑟𝑑 edition, John Wiley
and Sons, Inc. U A.

2. Moore D.S, McCabe G.P, Craig B. (2009) Introduction to the Practice of Statistics,
6𝑡ℎ 𝑒𝑑𝑖𝑡𝑖𝑜𝑛 ,W H Freeman and company, New York.

3. Montgomery C.D, Runger C.G. (2002) Applied Statistic and Probability for
Engineer’s, 3𝑟𝑑 edition, John Wiley and Sons, Inc, U.S.A.

4. Soong.T.T. (2004) Fundamental of Probability and Statistics for Engineers, John

Wiley and Sons, Inc, U.S.A.
 THE UNIVERSITY OF ZAMBIA
SCHOOL OF EDUCATION

NAME: RICHARD MOONDE

COMPUTER#: 14003635
COURSE CODE: EDU 3011
LECTURER: DR. PHIRI.W
DUE DATE: 14TH FEBRUARY, 2017

ASSIGNMENT 1