0% found this document useful (0 votes)
26 views13 pages

ASB Old Format

This document provides information about an exam for a course in applied statistics for business. The exam contains 50 multiple choice questions across 4 sections testing concepts related to descriptive statistics, inferential statistics, contingency tables, and paired sample tests. Additional information is provided on choosing an appropriate statistical test based on descriptive statistics.

Uploaded by

yutatyong127
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
26 views13 pages

ASB Old Format

This document provides information about an exam for a course in applied statistics for business. The exam contains 50 multiple choice questions across 4 sections testing concepts related to descriptive statistics, inferential statistics, contingency tables, and paired sample tests. Additional information is provided on choosing an appropriate statistical test based on descriptive statistics.

Uploaded by

yutatyong127
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

BEO2255 A pplied Statistics f or Business

PRACTICE EXAM PAPER

BEO 2255 Applied Statistics for Business


Student Number:

Seat Number:

Instructions
 This examination paper contains 50 questions.

 Each question carries equal marks (2 marks per question and a total of 100 marks. This will be
converted to 60 percent).

 Please number and write your answer on the answer booklet


provided. Answers should be precise and short. Detailed and lengthy
answers are not expected.

 Remember to write your name (family name first), ID number and seat number on both the exam
paper and answer book.

 Students are encouraged to show calculations and workings from your answers separately.


 DO NOT LEAVE THE EXAMINATION ROOM WITH THIS EXAM PAPER and
Handover both exam paper and answer book to invigilators upon the completion of
examination.

 ATTEMPT ALL QUESTIONS.

NOTE: This is based on the old syllabus, face-to-face exam.


You may want to try Session 8 Discriminant Analysis and
Section 9 Factor Analysis if you wish.

1
BEO2255 A pplied Statistics f or Business

Section 1
Section 1 may contain 15-17 questions that are similar in nature to the following questions.

1. What is the main advantage of an open question in a questionnaire compared to a closed


question?

2. When every possible element in the sample has the same chance of being selected, this sampling
method is called…………………

3. Name two errors that give rise to non-sampling errors.

4. You have collected a set of data on the attributes of successful and unsuccessful companies.
Name the statistical tool that can be applied to classify companies as successful and
unsuccessful?

5. What is the serial correlation (auto correlation) in regression analysis involving time-series data?

Section 2

Below is an SPSS table showing descriptive statistics on monthly food expenditure of white-collar
workers and blue-collar workers.

Group Statistics
Occupation N Mean Std. Std. Error
Deviation Mean
Expenditure on food White-collar worker 60 508.67 65.63 8.47
($ per month)
Blue-collar worker 60 503.02 65.27 8.43

17. Using descriptive statistics from the table above, estimate the 95 percent confidence intervals
of the mean food expenditure for white-collar workers and blue-collar workers in the target population.

18. Provide brief verbal statements to explain the meaning of the confidence interval estimates
obtained in Question 18 in the context of the data.

2
BEO2255 A pplied Statistics f or Business

Section 3
A random sample of individuals was asked to respond to the statement “mobile phones are harmful to
our health” (Attitude). They were also asked for their age group (Age). The results are as follows:

Atti tude * AGE Cros sta bula ti on

AGE
u nd e r 2 1 4 6 ye a rs
yea rs 2 1 - 4 5 yea rs & a b ove T ot a l
Atti tu d e Stro n g ly di sa g re e Cou n t 2 2
% wi th in Atti tu de 1 00 .0 % 1 00 .0 %
% wi th in AGE 2 .2 % 1 .0 %
% o f T ot al 1 .0 % 1 .0 %
Disag re e Cou n t 8 11 13 32
% wi th in Atti tu de 2 5. 0% 3 4. 4% 4 0. 6% 1 00 .0 %
% wi th in AGE 1 6. 3% 1 8. 3% 1 4. 3% 1 6. 0%
% o f T ot al 4 .0 % 5 .5 % 6 .5 % 1 6. 0%
No vi ew wi th er wa y Cou n t 19 16 25 60
% wi th in Atti tu de 3 1. 7% 2 6. 7% 4 1. 7% 1 00 .0 %
% wi th in AGE 3 8. 8% 2 6. 7% 2 7. 5% 3 0. 0%
% o f T ot al 9 .5 % 8 .0 % 1 2. 5% 3 0. 0%
Agre e Cou n t 15 23 35 73
% wi th in Atti tu de 2 0. 5% 3 1. 5% 4 7. 9% 1 00 .0 %
% wi th in AGE 3 0. 6% 3 8. 3% 3 8. 5% 3 6. 5%
% o f T ot al 7 .5 % 1 1. 5% 1 7. 5% 3 6. 5%
Stro n g ly ag re e Cou n t 7 10 16 33
% wi th in Atti tu de 2 1. 2% 3 0. 3% 4 8. 5% 1 00 .0 %
% wi th in AGE 1 4. 3% 1 6. 7% 1 7. 6% 1 6. 5%
% o f T ot al 3 .5 % 5 .0 % 8 .0 % 1 6. 5%
T ot a l Cou n t 49 60 91 2 00
% wi th in Atti tu de 2 4. 5% 3 0. 0% 4 5. 5% 1 00 .0 %
% wi th in AGE 1 00 .0 % 1 00 .0 % 1 00 .0 % 1 00 .0 %
% o f T ot al 2 4. 5% 3 0. 0% 4 5. 5% 1 00 .0 %

Chi-Square Tests

Asymp. Sig.
Value df (2-sided)

Pearson Chi-Square 5.272a 8 .728


Likelihood Ratio 5.960 8 .652
Linear-by-Linear .429 1 .512
Association
N of Valid Cases 200

a. 3 cells (20.0%) have expected count less than 5. The minimum expected count is .49.

19. How do you classify these two variables (“Attitude” and “Age”)?

20. For which specific test do you use contingency tables (results are given above)?

21. Complete the following statement by filling the blank.


percent of those respondents who disagree that mobile phones are harmful to our health
are between 21 to 45 years old.

22. If a person is selected randomly, what is the probability that he or she will be 46 years old or
older?

3
BEO2255 A pplied Statistics f or Business

23. At the 0.05 level of significance, based on results of Chi-square test, what is your decision and
conclusion (The test is whether the attitude and age are independent).

Section 4
ABC Company wants to find out whether its diet programme will result in weight loss. A random
sample of 100 people was subjected to the diet programme. Their weights (in kilograms) were taken
before and after undergoing the diet programme. Because ABC Company is uncertain whether to use a
paired samples t-test or Wilcoxon signed rank test for this study, it has produced the below descriptive
statistics table (Table A) to help select the appropriate statistical technique. Additionally Table B
bellow shows the results of the Wilcoxon signed rank test.

Table A: Descriptives
Descriptive Statistics
N Skewness Kurtosis

Statistic Statistic Std. Error Statistic Std. Error

Difference = Before - After 100 -3.586 .241 25.552 .478


Valid N (listwise) 100

Table B: Wilcoxon Signed Ranks Test


Ranks

N Mean Rank Sum of Ranks

After (kg) - Before (kg) Negative Ranks 57a 51.21 2919.00


Positive Ranks 34 b 37.26 1267.00
Ties 9c
Total 100

a After (kg) < Before (kg),b After (kg) > Before (kg)
c Before (kg) = After (kg)
Test Statistics
After (kg) - Before
(kg)

Z -3.371a
Asymp. Sig. (2-tailed) .001

a Based on positive ranks.


b W ilcoxon Signed Ranks Test

24. Use descriptive statistics (Table A) given above, to determine whether you can use a non-
parametric test (Wilcoxon signed rank test) as opposed to a classical t test (you should justify
your answer).

4
BEO2255 A pplied Statistics f or Business

25. In order to test whether there is no difference in weights before and after, you are requested to
write the null and the alternative hypothesis.

26. Based on the answers to question 25 and 26 above, at 0.05 level of significance, conduct the
hypothesis test and write down your conclusion (justify your answer).

Section 5
Supersales Company is interested to find out whether the distribution of salaries between its male and
female sales staff are the same. They decided to use the Wilcoxon rank sum test. The SPSS output is
as follows:
Ranks

Sex N Mean Rank Sum of Ranks


Salary ($’000) Male 50 51.59 2579.50
Female 50 49.41 2470.50
Total 100

Test Statistics a
Salary ($’000)
Mann-W hitney U 1195.500
Wilcoxon W 2470.500
Z -.376
Asymp. Sig. (2-tailed) .707
a Grouping Variable: Sex

27. In order to conduct a hypothesis test in comparing the distribution of salaries between male and
female staff of the company, you are requested to write the appropriate hypotheses.

28. What is your decision and conclusion at the 5 percent level of significance in the above
hypothesis test (Wilcoxsun rank sum test).

Section 6
A marketing study is conducted to investigate the effects of brand and shelf location on weekly biscuits
sales. Biscuits sales are recorded for each of three brands (Brand A, Brand B, and Brand C) and three
shelf locations (bottom, middle, and top). The following tables provide the results of One-way and
Two-way ANOVA tests and use them to answer following questions.

5
BEO2255 A pplied Statistics f or Business

ANOVA
SALES

Sum of Squares df Mean Square F Sig.


Between Groups 38219.550 2 19109.775 49.042 .000
Within Groups 45590.375 117 389.661
Total 83809.925 119

29. You are requested to write the appropriate hypotheses, to test whether the population mean sales
of biscuits are the same for all brands.

30. Using the statistics from the one-way ANOVA table above, conduct the above hypothesis test
and state your decision regarding the null hypothesis (test at 0.05 level of significance).

31. Following the Question number 30 and based on the hypothesis test conducted, write down your
conclusion.

Tests of Between-Subjects Effects

Dependent Variable: SALES


Source Type III Sum of
Squares df Mean Square F Sig.
Corrected Model 41927.118a 7 5989.588 16.017 .000
Intercept 2144372.779 1 2144372.779 5734.328 .000
LOCATION 1620.021 2 810.011 2.166 .119
BRAND 39898.629 2 19949.314 53.347 .000
LOCATION * BRAND 1372.121 3   
Error 41882.807 112
Total 2651375.000 120
Corrected Total 83809.925 119

a. R Squared = .500 (Adjusted R Squared = .469)

32. Using statistics from the above two-way ANOVA table (edited), test whether the population
mean sales of biscuits are the same for all locations and write down your decision and conclusion
(test at 0.05 level of significance).

33. Using statistics from the above two-way ANOVA table (edited), test whether the population
mean sales of biscuits are the same for all brands and write down your decision and conclusion
(test at 0.05 level of significance).

6
BEO2255 A pplied Statistics f or Business

Multiple Comparisons

Dependent Variable: SALES

Tukey HSD
95% Confidence Interval
Mean Difference
(I) Brand (J) Brand (I-J) Std. Error Sig. Lower Bound Upper Bound

Brand A Brand B -8.77 4.32 .110 -19.05 1.50


Brand C 32.70 4.32 .000 22.43 42.97
Brand B Brand A 8.77 4.32 .110 -1.50 19.05
Brand C 41.47 4.32 .000 31.20 51.75
Brand C Brand A -32.70 4.32 .000 -42.97 -22.43
Brand B -41.47 4.32 .000 -51.75 -31.20
Based on observed means.

34. Referring to the above table, at the 0.05 level of significance, identify a pair of brands with
statistically significant mean difference in sales of biscuits.

Section 7
A household cleaning product manufacturer wants to investigate the monthly household expenditure
on cleaning products. A survey was conducted on a random sample of purchase decision makers from
separate households. The variables are PURCHASE (monthly purchase in $), INCOME (yearly
household income in $’000), and location dummy variables (CITY = household located in the city
areas, URBAN = household located in the urban areas, RURAL = household located in the rural areas).
RURAL is considered as the base level. Some of the SPSS tables are as follows:

Descriptive Statistics

Mean Std. Deviation N


PURCHASE 486.9444 163.8918 90
INCOME 48.3222 10.5232 90
CITY .42 .50 90
URBAN .31 .47 90
a
Coefficients

Standardized
Unstandardized Coefficients Coefficients

Model B Std. Error Beta t Sig.


1 (Constant) 41.658 60.545 .688 .493
INCOME 7.446 1.305 .478 5.704 .000
CITY 145.881 33.537 .442 4.350 .000
URBAN 76.780 32.912 .218 2.333 .022
a Dependent Variable: PURCHASE

7
BEO2255 A pplied Statistics f or Business

35. Write the estimated regression equation.

36. Estimate the monthly household spending on cleaning products of a household located in a rural
area and has an annual income of $55,500?

Model Summary

Adjusted R Std. Error of the


Model R R Square Square Estimate
1 .713a .508 .491 116.9594
a Predictors: (Constant), INCOME, CITY, URBAN

37. What conclusion/conclusions you can make regarding the goodness of fit of the model (hint: use
Adj R2 value from the Model summary table)?

38. Calculate CV (Coefficient of Variation) and state the size of the regression error in terms of the
average monthly expenditure on cleaning products.

ANOVA b
Model Sum of Squares df Mean Square F Sig.
1 Regression 1214150.5 5 404716.839 29.586 .000a
Residual 1176436.2 86 13679.491
Total 2390586.7 89

a Predictors: (Constant), INCOME, CITY, URBAN, b Dependent Variable: PURCHASE

39. Using the results of ANOVA table above, test whether the regression model is significant (at
0.05 level of significance).

8
BEO2255 A pplied Statistics f or Business

Graph - Simple Scatterplot


30 0

20 0

10 0
Unstandardized Residual

-100

-200

-300
10 0 20 0 30 0 40 0 50 0 60 0 70 0 80 0 90 0

PURCHASE

40. Based on the above simple scatterplot, is the model hetroskedastic? Briefly explain your answer.

* Section 8
Discriminant analysis is used to examine the characteristics of gym members that make them more
likely to be members rather than non-members. The variables used in the function are the respondent’s
-
income in $’000 (INCOME), number of working - hours per week (WORKING), whether the respondent
owns a car or not (CAR = 1 if yes, 0 otherwise), whether the respondent is involved in other sport
activities (SPORT = 1 if yes, 0 otherwise), the gender of the respondent (SEX = 1 if male, 0 otherwise),
whether the respondent’s employment is classified as professional (PROFESS = 1 if yes, 0 otherwise).
The member variable (MEMBER) is coded 1 = yes and 0 = non-member. You are requested to use the
SPSS outputs given below to answer the following questions.

9
BEO2255 A pplied Statistics f or Business

Group Statistics
Valid N (listwise)
Std.
MEMBER Mean Deviation Unweighted Weighted
Non-member INCOME 43.71 4.58 63 63.000
WORKING 36.86 8.79 63 63.000
CAR .29 .46 63 63.000
SPORT .16 .37 63 63.000
SEX .71 .46 63 63.000
PROFESS .29 .46 63 63.000
Member INCOME 56.39 5.80 57 57.000
WORKING 36.86 10.03 57 57.000
CAR .60 .49 57 57.000
SPORT .40 .49 57 57.000
SEX .70 .46 57 57.000
PROFESS .84 .37 57 57.000
Total INCOME 49.73 8.19 120 120.000
WORKING 36.86 9.36 120 120.000
CAR .43 .50 120 120.000
SPORT .28 .45 120 120.000
SEX .71 .46 120 120.000
PROFESS .55 .50 120 120.000

41. Using the Group Statistics presented in the table above, identify 4 important predictor variables.
4
important predictors Income , ear ,
are
sport and Profess

Pooled Within-Groups Matrices

INCOME WORKING CAR SPORT SEX PROFESS


Correlation INCOME 1.000 &
.247 .004 .173 .133 .097
WORKING .247 1.000 .022 -.034 .037 .057
CAR .004 .022 1.000 -.189 -.106 -.076
SPORT .173 -.034 -.189 1.000 -.012 .036
SEX .133 .037 -.106 -.012 1.000 .020
PROFESS .097 .057 -.076 .036 .020 1.000

42. Based on the correlation coefficients shown in the table above (Pooled Within-Groups Matrices),
identify variables that might contain collinearity between the variables?

variables that contain collinearity


might is between
Income and

Working with the correlation coefficient of 0


247 .

10
BEO2255 A pplied Statistics f or Business

Eigenvalues
Canonical
Function Eigenvalue % of Variance Cumulative % Correlation
1 2.105 a 100.0 100.0 .823

a First 1 canonical discriminant functions were used in the analysis.

43. Acoording to the statistics shown in the Eigenvalues table above, what percentage of the &
&

variation in purchasing behaviour is explained by the discriminant function?


10 823)2
.
0 677 x 100
=
.
% 67 7 % = .

Wilks' Lambda
Test of Function(s) Wilks' Lambda Chi-square df Sig.
1 .322 130.285 6 .000

44. Use the statistics given in the Wilks’ Lambda table above, to test whether the discriminant
function is significant in differentiating between members and non-members at the 0.05 level of
significance.
&
Therefore , reject null hypothesis
Standardized Canonical Discriminant Function Coefficients
Function
1
INCOME .865
WORKING -.237
CAR .264
SPORT .072
SEX -.096
PROFESS .414

45. According to the Standardized Canonical Discriminant Function table above, name the most
important as well as the least important variables in the discriminant function?
most
important -

Income 10 865)
.

,
least important sport 10 072)
.

Functions at Group Centroids


Function
MEMBER 1
Non-member -1.368
Member 1.512
Unstandardized canonical discriminant functions evaluated at group means

? 46. Estimate the mid-point between the centroids using group centroids given in the above table.

( -

1 368 + 1
.

512)/2 0 072
=
.

-
7 2 % .

11
BEO2255 A pplied Statistics f or Business

Canonical Discriminant Function Coefficients


Function
D = -

8 .

0467 0 .
167 Income
-

0 .
025 Working + 0 .

557
1
Sex + 994 Profess
Car + 0 166 sport 0 208 0 .
-

INCOME .167
.

WORKING -.025 -

CAR .557
SPORT .166
SEX -.208
PROFESS .994
(Constant) -8.046
Unstandardized coefficients

47. Write the estimated discriminant function for calculating the discriminant score using
coefficients from the above table.

* Section 9
In an analysis of the results of a random sample survey concerning Yummy chocolates, buyers are
asked their responses to six propositions regarding the product on a 5-point Likert scale. The variables
and statements are as follows:

Variables Propositions
VALUE = ‘Yummy chocolates give value for money.’
ATTRACT = ‘Yummy chocolates are packed in attractive boxes.’
SMOOTH = ‘Yummy chocolates are smooth in taste.’
PRICE = ‘Yummy chocolates are not expensive.’
NUTRI = ‘Yummy chocolates are nutritious.’
SIZE = ‘Yummy chocolates come in convenient pack sizes.’
CREAMY = ‘Yummy chocolates are creamy in taste.’

12
BEO2255 A pplied Statistics f or Business

The factor analysis shown in the following tables used principal components, orthogonal rotation and
varimax.

Total Variance Explained


Initial Eigenvalues Rotation Sums of Squared Loadings

Component Total % of Variance Cumulative % Total % of Variance Cumulative %

1 2.655 37.925 37.925 2.557 36.527 36.527


2 1.604 22.919 60.845 1.630 23.279 59.806
3 1.469 20.989 81.833
&
1.542 22.027 81.833
4 .446 6.367 88.200
5 .309 4.415 92.615
6 .292 4.170 96.785
7 .225 3.215 100.000

Extraction Method: Principal Component Analysis.

48. Based on the Total Variance table above, what is the percentage of variation that the first three
factors explain in the responses to the questions?
81 83/
.
&

Rotated Component Matrix a


Component
1 2 3
VALUE -.116 .895 -5.271E-02
ATTRACT -.135 -3.332E-02 .900
SMOOTH .915 -4.838E-02 8.341E-02
PRICE 2.803E-02 .906 6.849E-03
NUTRI .896 -6.038E-02 -1.941E-02
SIZE .280 -1.267E-02 .846
CREAMY .898 -8.651E-03 8.052E-02

Extraction Method: Principal Component Analysis.


Rotation Method: Varimax with Kaiser Normalization.
a.
Rotation converged in 4 iterations.

49. How many factors are identified in the Rotated Component Matrix?
3 factors are identified

50. According to the Rotated Component Matrix what are the variables that are heavily loaded onto
factor 1?

Hearing loaded onto Factor I are Nutri and


Creamy + Smooth 13

You might also like