ASSIGNMENT

MAY 2020 SEMESTER

SUBJECT CODE : ESE633

SUBJECT TITLE : STATISTICS IN EDUCATION

LEVEL : MASTER

STUDENT’S NAME :

MATRIC NO. :

PROGRAMME :

ACADEMIC :
FACILITATOR

LEARNING CENTRE :
Question 1: [3 marks]

Class X

Class Y

An English test was

administered to two classes of secondary school students. Explain the distribution of the scores
and the standard deviation shown in the two graphs above.

Answer:

An English test was administered to Class X and Class Y and the distribution of
the scores are shown below:
 In Class Y, the scores are widely spread out which means that there is high variance or a
bigger standard deviation which most the scores are in red color. If the mean is 50, then
we can say that 95% of the students scored between 44 and 56.
 In Class X, there is low variance or a small standard deviation which explains why most
of the scores are clustered around the mean. Most of the scores are 'bunching' around the
mean; and shown that most of the scores in red color. If the mean is 50, 95% of the
students scored between 47 and 53.
Question 2 [3 marks]

Gender Male Female

Reading Score 48.8 49.1
Standard Deviation 3.56 4.46

Refer to the table above. Interpret the mean and standard deviation.

Answer:

The female score were little over 0.3 greater on average than the male score. The female having
an average of 49.1 and the male having an average of 48.8. The female data had standard
deviation of 4.46 whereas the male data had 3.56. The female data having a higher standard
deviation tells me that the female data is more spread out or dispersed than male data.

Question 3: [5 marks]

A research was conducted on the opinion of teachers on students bringing smart phones to
school. A sample 200 female and 165 male teachers were asked the question ‘Should students be
allowed to bring smart phones to school. The results of the study are shown below:

Should students be allowed to bring Total

smartphones to school?
Gender No Yes
Female 31 169 200
Male 15 150 165
Total 46 319

2
The χ value is 20.704 while the p-value is 0.0001.

Based on the information shown in the table:

a) Write a suitable null and the alternate hypotheses.

b) Why was the chi square used?
c) What conclusion can be drawn from the table above?
Answer:
a) From my study, determine that there is no difference by allowed to bring smartphones to
school between males and females; So. null hypothesis Ho: µ1 = µ2. The alternative
hypothesis Ha: µ1 ≠ µ2. A hypothesis test whose alternative hypothesis has this
form is called a TWO-TAILED TEST.
b) A chi square (X ) statistic is used to investigate whether distributions of
2

categorical variables differ from one another. The Chi Square statistic
compares the tallies or counts of categorical responses between two
independent groups Male and Female
c) Both male and female agree students be allowed to bring smartphones to school. At =
0.05, the data does provide sufficient evidence to conclude that the mean scores on
allowed student smartphones to school of females is superior to that to males, even
though the mean scores obtained is higher than that of males

Question 4: [5 marks]

Refer to the graph below and answer the following questions:

Reading
Score

Hours spent playing video games

a) Explain the relationship between the two variables

b) State the statistical test that can be used to identify the relationship
Answer:
a) Negative Correlation. An inverse linear relationship between Hours spent playing video
games and Reading Score.
 When Hours spent playing video games values are higher related to Reading
Score values are lower
 When Hours spent playing video games values are lower related to Reading Score
values are higher.
b) A chi-square test is used when you want to see if there is a relationship between two
categorical variables.

Question 5: [5 marks]

A physical education teacher collected data on students’ height AND the span of jump (see Table 1
below:

Student Height (m) Span of jump (m)

Table 1: 1 1.63 2.34
2 1.80 2.48
3 1.75 2.29
4 1.86 2.62
5 1.83 2.64
6 1.71 2.30
7 1.75 2.44
8 1.96 2.67
9 1.60 2.39
10 1.68 2.47
11 1.80 2.60
12 1.87 2.75
13 1.74 2.40
14 1.67 2.46
15 1.64 2.33

a) What statistical procedure would you use to predict length of jump using height? Give reasons.
b) Explain clearly two assumptions that must be met in conducting the test that you have identified
in Q1(a). Why it is important these assumptions are met?
Answer:

a) Linear Regression is a Statistical Method that use to test the extent to which causal
relationships exist between the height (X) and span of jump(Y). Causal Factors are
generally denoted by X or referred to as Predictors while Affect Variables are denoted by
Y or otherwise called Response. Simple or often abbreviated Linear Regression also one
of the Statistical Methods used in production to make predictions or predictions of quality
and quantity characteristics.
b) Based on above data conduct Linear Regression Analysis for two assumptions:
I. Test the relationship between height and Span of jump. Expecting a positive
relationship between height and span of jump. In other words, as height increases,
you expect span of jump to also increase. How do you establish this to be true?
II. Second, have go further than just stating the relationship between height towards
physical and span of jump. You want to know whether you can PREDICT
values of one variable if you know or can estimate the other variable. In other
words, can predict performance in span of jump on what you know about their
height towards physical.

Question 6: [4 marks]

Critical Thinking N
High Income 54.2 120
Middle Income 60.1 120
Low Income 48.2 120

A study was conducted to test the critical thinking abilities of primary school children from different
socioeconomic backgrounds. The results are shown in the table above. Explain why you would use
Oneway ANOVA and not multiple t-tests to determine if the difference between the groups is significant.
Answer:
T-test was used to compare differences of means between two groups, such as comparing
outcomes between a control and treatment group in an experimental study. But in this case, by
multiple t-tests enhances the likelihood of committing Type 1 which can claim that two means
are not equal when in fact they are equal. It does also reject a null hypothesis when it is TRUE.
On a practical level, using the t-test to compare many means is a cumbersome process in
terms of the calculations involved.

Question 7: [7 marks]

A researcher conducted a study to measure Reasoning Skills among a group of 16- year
students. The sample consisted of 120 males and 126 female subjects. Reasoning Skills
consists of two constructs – Inductive Reasoning and Deductive Reasoning.

a) Suggest THREE possible research questions.

b) State the appropriate statistical tests to test the three research questions suggested.

Answer:
a) 1. the extent to which students in the 16-year-olds master the reasoning skills
2. male students master more reasoning skills than female students
3. female students master more reasoning skills compre than male

b) I recommend using the T Test for statistical tests to test the three research questions suggested.
Used when individual in the sample was measured twice and both measurement data used to create
comparison. Paired sample t tests is used when two sets of data are present from just one group same
subject (1 sample).
Question 8 (15 marks)
A study was conducted to compare the effectiveness of three teaching techniques in teaching English to
Year 6 students. The three teaching techniques are – Using video clips, Using worksheets and Using flash
cards. Equal number of students were randomly assigned to the three different treatments. The results are
shown in the tables below:

Scores in Statistics
Treatment N Mean Std. Deviation Std. Error

Using Video Clips 20 27.20 3.05 .96

Using Worksheets 20 23.60 3.31 1.04
Using Flash Cards 20 23.40 3.24 1.02
a) Describe the data shown in the table above.
b) What does standard error in the table above indicate?

Test of Homogeneity of Variances

Statistics Score
Levene df1 df2 Sig.
Statistic
.115 2 57 .802

c) Interpret the Levene Statistic in the table above.

ANOVA
Statistics Score

Sum of df Mean F Sig.

Squares Square
Between Groups 91.467 2 45.733 4.467 .031
Within Groups 276.400 57 10.237
Total 367.867 59

b) What can you conclude from the ANOVA table above

Dependent Variable: Statistics Score

 Tukey HSD
(I) Treatment (J) Treatment Mean Std. Error Sig.
Difference
(I-J)
Lecture with video Lecture only 3.600 1.431 .046
clips Small group 3.800 1.431 .034
Lecture only Lecture with video -3.600 1.431 .046
clips
Small Group .2000 1.431 .989
Small group Lecture with video -3.800 1.431 .034
clips
Lecture only -.2000 1.431s .989

Explain the Tukey HSD results in the table above.

Answer :
Question 8 a:

1. The Mean explain about Using Video Clips have the highest mean (27.02), while Using Flash
Cards have the lowest mean (23.40). Using Worksheets fall in the middle, with a mean of
23.60.

2. Standard Deviations explain that the standard deviation for Using Video Clips (3.05) and
Using Flash Cards (3.24) students is fairly close while Using Worksheets have a somewhat
bigger standard deviation of 3.31

3. Refer to the table, the three Standard Errors shown that 0.96 for Using Video Clip.
However, the standard error for Using Worksheets and Using Flash Cards are comparatively
high = 1.04 and 1.02.

Question 8 b:
The standard error is a measure of how much the sample means vary if you were to take repeated
samples from the same population. The treatment using worksheet and using flash card contain
20 students each; the standard error of the mean for each of the last two groups is high. It is 1.04
for Using Worksheet and 1.02 for Using Flash Card. However, the standard error for the Using
Video clips group is comparatively low = 0.96. It because the smaller number of low ability
students and the larger standard deviation explains why the standard error is larger.

Question 8C:
The Levene's test of homogeneity of variance is used for the OneWay ANOVA and is show in
the Table above. The p-value which is 0.802 is greater than the alpha of 0.05. Hence, it can be
concluded that the variances are homogeneous which is reported as Levene (2, 57) = 0.115, p =
0.802

Question 8 d):
Having concluded that the assumption of homogeneity of variance has been, the means and
standard deviations of each of the four groups have been computed. The table above indicates as
below:-
 The ‘Between groups’ row shows that the df is 2 (i.e. k - 1 = 3 – 1 = 2) and the mean
square is 45.733.
 The ‘Within groups’ row shows that the df is 57 (N - k = 60– 3 = 57) and the mean
square is 10.237.
 If you divide 45.733 by 10.237 you will get the F value of 4.467 which is significant at
0.031.
 Since, 0.031 is < than α = 0.05, we can Reject the Null Hypothesis and accept the
alternative hypothesis. So. It concludes that there is a significant difference in inductive
reasoning between the four SES groups.
Question 8E:
There is a significant difference ONLY between ‘Using Video Clips’ (Mean = 27.20) and “Using
Flash Cards (Mean = 23.40) at p = 0.034. The Using Video Clips scored significantly higher than
Using Flash Cards at p = 0.034. There are no significant differences between the other groups.
Question 9 [7 marks]
 A study was conducted to determine the relationships between spatial reasoning, memory,
metacognition, mathematical ability and verbal reasoning. The sample consisted of 80 male and
80 female 16 year old students. The Table above shows a correlation matrix between the five
variables. Based on the table answer the following questions:
Spatial Memory Metacogniton Mathematical Verbal
Reasoning ability reasoning

Spatial reasoning
Memory 0.56
Metacognition 0,65 0.67
Mathematical ability 0.43 0.60 0.59 .
Verbal reasoning 0.34 0.41 0.49 0.70

a) State THREE research questions

b) Discuss the findings of the study.

Answer: a)
i. There is an association between Spatial reasoning with Memory factors
ii. There is an association between Spatial reasoning with Verbal reasoning factors
iii. There is an association between Spatial reasoning with Metacognition factors

Answer b):
i. The results of this hypothesis clearly prove that there is a relationship significant positive
among variables of Spatial reasoning with Memory. The value of, r is 0.56. Which is less
than 0.05 (r = 0.865, p <.05). This means there is a link between Spatial reasoning with
Memory factors.
ii. In the research done, it is clear that there is a relevance significant positive correlation
between spatial reasoning and Verbal reasoning factor. This is shown by the lower
correlation value of 0.34 which is close to +1, while the significance value, p is smaller than
0.05, which is 0.012.
iii. There is a connection significant difference between Spatial reasoning and Metacognition
factors. Numerically, we can see the correlation coefficient value, that is 0.65 and the
significance level is high than 0.05. (r = .65, p >.05).
iv. There is a connection significant difference between Spatial reasoning and Mathematical
ability factors. Numerically, we can see the correlation coefficient value, that is 0.43 and the
significance level is less than 0.05. (r = .43, p <.05).

Question 10: [6 marks]

Table 1:
Gender Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig. See
Table
Spatial Male 0.879 34 0.012 0.971 78 0.016 above
Test Female 0.185 39 0.001 0.079 96 0.008 and
Scores answer
the
a. Lilliefors Significance Correction following:

a) What is the purpose of the statistical test used?

b) Explain the difference between ‘Kolmogorov-Smirnov’ and ‘Shapiro-Wilk’.

c) What can you conclude?

Answer :

a) The purpose of the statistical test is to provide a mechanism for making quantitative
decisions about a process or processes. The intent is to determine whether there is enough
evidence to "reject" a conjecture or hypothesis about the process.
b) The Shapiro-Wilk used fairly powerful omnibus test. It’s also not good with small samples
or discrete data. Its Good power with symmetrical, short and long tails. Good with
asymmetry. In other hand, Kolmogorov-Smirnova used in all tends to have lower power.
Data have to be very non-normal to reject Ho., and these tests can outperform other tests
when using discrete or grouped data.

Question 11: [6 marks]

A researcher was interested in finding out whether Moral Reasoning could be enhanced if
students were taught using Moral Dilemmas.Subjects were given a Moral Reasoning Test
before the treatment (using moral dilemmas) and after the treatment and the results are shown
in Table 1 below.

Table 1 : Mean Moral Reasoning Score Before and After

Teaching Using Moral Dilemmas
N Mean Std. Std. Error Mean
Deviation
Pretest 30 18.50 5.33 0.97

Posttest 30 23.86 4.75 0.87

a) Describe the findings in Table 1

b) State the null hypothesis
c) State the alternative hypothesis

Table 2 : Paired t Test

Mean Std. Deviation Std. Error Lowe Uppe t df Sig.
Mean r r (2 tailed)
Pretest -5.36 2.90 0.62 -6.65 -4.08 - 29 .000
8.66
Posttes
t
d) Why is the paired t-test used? [as shown in Table 2 above]
e) What can you conclude from Table 2?

Answer a):

 According the table the Mean shown that, Posttest has the highest
mean (23.86), while Pretest has the lowest mean (18.50). The Standard Deviations tell us
about standard deviation for Pretest (5.33) and Posttest (4.75). This is the Standard Deviation
of Pretest higher than Posttest. Standard Errors refer to the table; the standard error is a
measure of how much the sample means vary if you were to take repeated samples from the
same population. The both groups contain 30 samples each, the standard error of the mean
for each of these groups are 0.97 for Pretest and 0.87 for Posttest.

b) There is no significant difference between the pretest and the posttest scores Moral
Reasoning Score Before and After Teaching Using Moral Dilemmas
c) Ha: µ1 ≠ µ2
Ha: The Alternative Hypothesis might be that the reasoning scores between
discovery pretest and posttest are DIFFERENT.
Ha: µ1 > µ2
Ha: The Alternative Hypothesis might be that the reasoning scores of the
discovery pretest is HIGHER than the mean scores of the posttest.
Ha: µ1 < µ2
Ha: The Alternative Hypothesis might be that the reasoning scores of the
discovery pretest is LOWER than the mean scores of the posttest.
d) The Paired Samples t Test can only compare the means for two related units on a
continuous outcome that is normally distributed. The Paired Samples t Test is not
appropriate for analyses involving the following:
i. unpaired data;
ii. comparisons between more than two units/groups;
iii. a continuous outcome that is not normally distributed; and
iv. an ordinal/ranked outcome.

e) Table above reports that the mean values on the variable for the pretest and posttest.
The posttest mean is higher (23.86) than the pretest mean (18.50) indicating improved
performance in the history test after the treatment. The standard deviation for the
pretest 5.33 and is very close to the standard deviation for the posttest which is 4.75.
This is the difference between the means 18.50 – 23.86 = – 5.36 which students did
significantly better on the posttest.
Question 12: [5 marks]

Parents Not Divorced 0.37,0.44 Parents Divorced 0.74,0.79

N Mean Std. N Mean Std. Sign
Deviation Deviation

Score on Anxiety 49 21.49 2.6 34 22.70 4.3 0.07

Test
Score on 49 15.67 3.1 34 17.90 4.6 0.03
Depression Test

The table above shows the results of a study which compared the anxiety and depression levels
of students from families whose parents are divorced and whose parents are not divorced. Briefly
describe the results of the study.

Answer|:

The results of the analysis showed that the mean and standard deviation of the Anxiety Test for
Parents Not Divorced were 21.49 and 2.6 respectively, while the mean and standard deviation of
the Anxiety Test for Parents Divorced were 22.70 and 4.3, respectively. The mean score
difference between the tests was -1.21. The results of the study were significant (t = −3.27, p
<0.07) This indicates that there is an increase in anxiety for divorced parents compared to not
divorced. Meanwhile, the mean and standard deviation of the Depression Test for Parents Not
Divorced were 15.67 and 3.1, respectively, and the mean and standard deviation of the Test for
Parents Divorced test were 17.90 and 4.6, respectively. The mean score difference between the
tests was -2.23. The results of the study were significant (t = −3.01, p <0.07) this indicates that
there is an increase in pressure for divorced parents compared to not divorcing
Question 13: [7 marks]

A researcher conducted a study to measure the Emotional

Intelligence of a group of 16-year old students. The sample
consisted of 120 subjects; 60 males and 60 female subjects. In her
study, the researcher defined Emotional Intelligence as consisting
of three factors or constructs; namely, Stress Tolerance,
Optimism and Emotional Self-awareness.

a) List THREE possible hypotheses for the study above [3 marks]

b) State the appropriate statistical tests to test the three hypotheses listed in (a)
[2 marks]

c) State the assumptions required for the statistical test(s) used in (b) [2 marks]
Answer:
a) i. There is no significant difference between the Emotional Intelligence and
constructs. (Null Hypothesis)
ii. The Alternative Hypothesis might be that the reasoning scores between
Emotional Intelligence and posttest are Stress Tolerance.
iii. The Alternative Hypothesis might be that the reasoning scores of the
Emotional Intelligence is HIGHER than the mean scores of the posttest
b) the appropriate statistical tests to test the three hypotheses T – test
c) As a parametric procedure the paired sample t-test makes several assumptions.
Although t-tests are quite robust, it is good practice to evaluate the degree of deviation
from these assumptions in order to assess the quality of the results. In a paired sample t-
test, the observations are defined as the differences between two sets of values, and each
assumption refers to these differences, not the original data values. The paired sample t-
test has four main assumptions:
 The dependent variable must be continuous (interval/ratio).
 The observations are independent of one another.
 The dependent variable should be approximately normally distributed.
 The dependent variable should not contain any outliers.
 Question 14: [5 marks]

Explain the meaning of the TWO graphs below.

Weight

Height [Berat]

Income Height [cm]

Answer:

I. ZERO CORRELATION
If Income (x) and Height (y) have NO relationship than the Slope
(β1) will be ZERO. In other words, there is NO SYSTEMATIC
RELATIONSHIP between X and Y. Some students with high Income have
positive low Height while some have low Income score have high
positive height.

II. POSITIVE CORRELATION

If Height (x) and weight (y) had a positive relationship than the Slope (β1) will be a
positive number. Lines with positive slopes go from the bottom left toward the upper
right.
Question 15: [11 marks]

A researcher conducted a study skills course over a period of five days among a group of
beginners, intermediate and advanced speakers of English. At the end of the programme he
administered a test to determine which group of students benefited from the programme.

The results of the study were analyzed using One-Way ANOVA and the results are shown in the
tables below:

Table 1

a) See Table 1 and answer the following questions:

i) State one research question for the study

ii) Why was the Oneway-ANOVA used?

iii) What can conclude from the above table?

iv) State the assumptions for using the One-way ANOVA.

b) See Table 2above and answer the following questions:

v) What can you conclude from the above table?

vi) What does the standard deviation tell you about the results?

vii) Explain the meaning of “95% confidence interval for mean”?

 c) See Table 3 above and answer the following questions:

viii) What is the meaning of “multiple comparison”?

ix) What can you conclude from Table 2?

x) Answer the Research Question stated in a (i)?

Answer (a):

I. The null hypothesis (H0) is that there is no difference between the groups and English
between means. (Walruses weigh the same in different months)