Research in Organisation and Business Context

Practical 2 and 3

Instructions

For each of the questions in practical 2 and 3 you are required to produce between 2 and 3
tables displaying your results i.e., at least the appropriate descriptive statistics table, the main
results table, and the homogeneity of variance table where necessary. Change the font of the
tables to Times New Roman size 9. Examples of these can be found at the end of this document.

Practical 2

For practical 2 you need to read the case studies below and answer the questions related to it:

Case study 1

Within the work context the association between perceived negative aspects of the work
environment and physical and psychological strain has been well documented. In 2015 a study
conducted by a South African University has reported that South African academics experience
higher levels of burn-out than academics in Europe. The report suggested that the higher levels
of burn-out were related to the increase in work role conflict experienced by South African
academics. The (then out of favour) Minister of Higher Education and Training, Dr.
Bonginkosi "Blade" Nzimande, commissioned a consulting company Research Limited to
investigate this phenomenon. Dr Nzimande was of the opinion that burn-out has nothing to do
with work role conflict. He argued that South African academics experience burn-out because
they are “overstaying their time at universities.” He instructed the researchers to investigate if
there is a difference in burn-out between academics who are working at a university for less
than 10 years (lecturers) and those working at the university for more than 10 year (senior
lecturers).

Question 1

1.1 What are the dependent and independent variables in Dr Nzimande’s hypothesis? (2)
DV= Burn-out

IV= Time spent at university

1.2 State the null and alternative hypothesis for Dr Nzimande’s premise. (2)
There is no difference in the time spent (< 10 yrs vs < 10 yrs) at university and
burnout scores/experienced.
 H0 = μ1 = μ2

H1 = μ1 ≠ μ2
1.3 State the design requirements for this study. (4)
There are two independent groups i.e., lecturers and senior lecturers.

There is one continuous variable on which the two groups are being compared i.e.,
burnout .

Random sampling

Normal distribution of score

Homogeneity of variance
1.4 State the decision rule for this study. (1)
Reject H₀ if ɑ ≤ 0.05

Do not reject H₀ if ɑ > 0.05

1.5 Report the t-statistic and make a decision about the null hypothesis. (3)
Descriptives

YearsUni BurnOut

N Lecturer 699
SeniorLecturer 398

Missing Lecturer 0
SeniorLecturer 0

Mean Lecturer 39.1

SeniorLecturer 39.8

Standard deviation Lecturer 4.92

SeniorLecturer 4.99

Minimum Lecturer 9
SeniorLecturer 9

Maximum Lecturer 45
SeniorLecturer 45

Homogeneity of Variances Tests

F df df2 p

BurnOut Levene's 0.578 1 1095 0.447

Variance ratio 0.974 698 397 0.763

Note. Additional results provided by moretests

Homogeneity of Variances Tests

F df df2 p

Independent Samples T-Test

Statistic df p Effect Size

BurnOut Student's t -2.19 1095 0.028 Cohen's d -0.138

t = -2.19

1.6 Explain in detail why the Independent Samples t-Test/ Two-Sample Assuming Equal
Variances was used. (7)

There are two examples of the two-sample t-test in excel: Two-Sample Assuming
Equal Variances and Two-Sample Assuming Unequal Variances. To decide between
the two (and because there are two different groups) we need to conduct an F-test.
The F-test is the homogeneity of variance test to determine if the two groups come
from the same population. If the p-value for the F-test is > 0.05, then we assume equal
variance and proceed with the Two-Sample Assuming Equal Variances. If the p-value
for the F-test is < 0.05, then we assume unequal variance and proceed with the Two-
Sample Assuming Unequal Variances.

1.7 What conclusions can be drawn from the results? Report on all the important values, there

implications and what they mean for Dr Nzimande’s hypothesis. (6)

An independent samples/two-sample/ t-test: two-sample assuming equal variances was

conducted to test if levels of burn-out experienced by academics differed between
lecturers and senior lecturers. The test showed a significant difference in the levels of
burn-out amongst lecturers (M = 39.1) and senior lecturers (M = 39.8), t (1095) = -2.19,
p < .05. Senior lecturers do indeed experience higher levels of burnout, with a marginal
or small effect as indicated by Cohen’s D = -0.138. (Reject the null hypothesis).
(25)
Case study 2

As a psychologist himself, Dr Nzimande is alarmed by a current Mail and Guardian article

reporting that because of the COVID-19 pandemic the levels of motivation amongst academics

are at its lowest in over 30 years. He commissions a consulting company to implement a

programme focused on developing coping skills as a strategy to increasing the coping and

motivation levels of academics. Measures of Managerial Skills, Employee Motivation and the

pre- and post-intervention Coping scores were obtained, as well as several organisational

outcome measures. These outcome measures were assessments of both personal/individual and

organisational level performance. Using the appropriate variables determine whether the

programme that was implement was an effective strategy for improving academics coping.

Question 2

2.1 What type of design and method of data collection was used in this study? 2

One group pre- and post-test design, self-completion questionnaire

2.2 What are the dependent and independent variables? 2

There will be no difference in coping skills amongst lecturers and senior lecturers
after attending a coping skills intervention.
DV=Coping skills
IV = Program implemented

2.3.1 State the hypothesis 1

H0 = μ1 = μ2 H0 = μ = 0

H1 = μ1 ≠ μ2 H1 = μ ≠ 0

2.3.2 State the design requirements and assumptions 3

There is a set of continuous scores representing the sample

There is a hypothetical distribution of scores with which the sample is being
compared
Sample was randomly drawn from the population
 Scores are normally distributed

2.3.3 State the decision rule 1

Reject H₀ if ɑ ≤ 0.05

Do not reject H₀ if ɑ > 0.05

2.3.4 Find and report the observed value (i.e. conduct the analysis) 8

Descriptives

N Mean Median SD SE

CopingPre 1097 76.1 80 19.03 0.575

CopingPost 1097 21.2 21 3.31 0.100

Paired Samples T-Test

Mean SE Effect
statistic df p
difference difference Size

Student's Cohen's
CopingPre CopingPost 101 1096 < .001 54.9 0.544 3.05
t d

2.3.5 Find and report the alpha level 1

ɑ = 0.01, critical 1.96 (2-tailed)

2.3.6 Make decision about H₀ 1

Reject H0

2.3.7 Follow the standard practice for writing and reporting the finding interpret and report the

results of the study, include the effect size. 6

A paired samples t-test was conducted to determine if the coping intervention

improved coping skills. The mean coping scores for academics before they took part
in the intervention was 76.1 (std dev = 19.03), whereas the post intervention mean
coping scores were 21.2 (std dev = 3.31). The independent samples t-test showed that
 the mean difference of 54.9 between the pre- and post-coping scores was significant
(t(1095) = 101, p < .001, Cohen’s D = 3.05).
We reject the null hypothesis (p < 0.05), hence there is difference in coping scores.
If high scores indicate good coping then the intervention was not successful, if low
scores indicate good coping then the intervention was successful. Thus depending on
whether the coping scores was positively or negatively scored would indicate whether
the intervention was successful or not.

(25)
[50]

___________________________________________________________________________

Practical 3

For practical 3 you need to follow the appropriate steps from practical 2 to be able to answer
the two questions of practical 3. The questions are related to the same case studies and data set
as in practical 2.

Question 1
Determine if there is a significant difference in managerial skills between male and female
academics. 25
3.1.1 State the hypothesis 1

There is no difference in managerial skills between male and female academics.

H0 = μ1 = μ2

H 1 = μ1 ≠ μ2
3.1.2 State the design requirements and assumptions 3

There are two independent groups

There is one continuous variable on which the two groups are being compared
Random sampling
Normal distribution of score
Homogeneity of variance

3.1.3 State the decision rule 1

Reject H₀ if ɑ ≤ 0.05

Do not reject H₀ if ɑ > 0.05

3.1.4 Find and report the observed value (i.e. conduct the analysis) 8

Descriptives

Gender ManagerialSkill

N Female 587
Male 510

Missing Female 0
Male 0

Mean Female 43.7

Male 43.7

Standard deviation Female 5.97

Male 6.13

Minimum Female 15
Male 22

Maximum Female 55
Male 55

Homogeneity of Variances Tests

F df df2 p

ManagerialSkill Levene's 3.02e-5 1 1095 0.996

Variance ratio 0.951 586 509 0.560

Note. Additional results provided by moretests

Independent Samples T-Test

Statistic df p Effect Size

ManagerialSkill Student's t -0.0176 1095 0.986 Cohen's d -0.00107

3.1.5 Find and report the alpha level 1

a = 0.05, critical 1.96 (2-tailed)

3.1.6 Make decision about H₀ 1

Fail to reject H0

3.1.7 Follow the standard practice for writing and reporting the finding interpret and report the

results of the study, include the effect size. 6

 An independent samples t-test was conducted to determine if male and female
academics differed in terms of their managerial skills. The mean managerial skills
scores for the male sample was 510 (std dev = 6.13), whereas the mean managerial
skills scores for the female sample was 587 (std dev = 5.97).
The t-test showed that there was no difference in managerial skills between male and
female academics (t (1095) = -0.02, p = 0.986). We therefore fail to reject the null
(p > .05).

Question 2
Determine if the programme implemented was successful as an intervention to improve
academics levels of motivation. 25
3.2.1 State the hypothesis 1

There is no difference in the levels of motivation between lecturers and senior

lecturers after attending the intervention.

H0 = μ1 = μ2 H0 = μ = 0

H1 = μ1 ≠ μ2 H1 = μ ≠ 0

3.2.2 State the design requirements and assumptions 3

There is a set of continuous scores representing the sample

There is a hypothetical distribution of scores with which the sample is being
compared
Sample was randomly drawn from the population
Scores are normally distributed
3.2.3 State the decision rule 1

Reject H₀ if ɑ ≤ 0.05

Do not reject H₀ if ɑ > 0.05

3.2.4. Find and report the observed value (i.e. conduct the analysis) 8

Descriptives

N Mean Median SD SE

PreMotivation 1097 7.45 8 1.71 0.0516

PostMotivation 1097 8.54 8 1.32 0.0399
Paired Samples T-Test

Mean SE Effect
statistic df p
difference difference Size

Student's Cohen's -
PreMotivation PostMotivation -21.2 1096 < .001 -1.08 0.0511
t d 0.640

3.2.5 Find and report the alpha level 1

a = 0.01, critical 1.96 (2-tailed)

3.2.6 Make decision about H₀ 1

Reject H0

3.2.7 Follow the standard practice for writing and reporting the finding interpret and report the

results of the study, include the effect size. 6

A paired samples t-test was conducted to determine if the intervention improved

academics level of motivation. The mean motivation scores for academics before they
took part in the intervention was 7.45 (std dev = 1.71), whereas the post intervention
mean motivation scores were 8.54 (std dev = 1.32).
The t-test showed that the mean difference of -1.08 between the pre- and post-
motivation scores was significant (t (1096) = -21.2, p < .001, Cohen’s D = 0.64).
We reject the null hypothesis (p < 0.05), hence there is difference in motivation scores.
This suggests that the intervention was successful in improving the motivation levels
of academics.

[50]
Practical 2 excel tables

Question 1 tables

F-Test Two-Sample for

Variances

Lecturer Snr Lecturer

Mean 39.11 39.79
Variance 24.24 24.88
Observations 699 398
df 698 397
F 0.97
P(F<=f) one-tail 0.38
F Critical one-tail 0.87

t-Test: Two-Sample Assuming Equal Variances

Snr
Lecturer Lecturer
Mean 39.11 39.79
Variance 24.24 24.88
Observations 699.00 398
Pooled Variance 24.47
Hypothesized Mean Difference 0.00
df 1095.00
t Stat -2.19
P(T<=t) one-tail 0.01
t Critical one-tail 1.65
P(T<=t) two-tail 0.03
t Critical two-tail 1.96
Practical 3

Question 1

F-Test Two-Sample for Variances

Female Male
Mean 43.67 43.67
Variance 35.70 37.52
Observations 587 510
df 586 509
F 0.95
P(F<=f) one-tail 0.28
F Critical one-tail 0.87

t-Test: Two-Sample Assuming Equal Variances

Female Male
Mean 43.67 43.67
Variance 35.70 37.52
Observations 587 510
Pooled Variance 36.54
Hypothesized Mean Difference 0
df 1095
t Stat -0.02
P(T<=t) one-tail 0.49
t Critical one-tail 1.65
P(T<=t) two-tail 0.99
t Critical two-tail 1.96