Statistics 1 Assignment 3

Salzburg University of Applied Sciences
Master Degree Program "Innovation and Management in Tourism"
Statistics 1
Assignment 3
Student Name: Catalina Martinez
Student numbers: 1910649025
Submission Date: 21.01.2020
Course Lecturer: Dimitri Prandner

ASSIGNMENT 2 Statistics 1
This assignment concerns itself with differentiation testing.

Thus, we will have to solve several problems that deal with differences between groups.
Each test is worth 2,5% of the mark on the assignment.
Part 1 T-Test and A Nova:
Q: Is there a significant difference between the sexes when it comes to their age?
USE: Age_2016, SEX_2016

Complete the following tasks:
1. Provide a univariate description (NO TABLES!) of the two variables.
● Regarding the Variable SEX_2016 the total number of respondents was n=2021, in
which 936 were male, representing 46,3% of the sample, and 1085 were female,
representing 53,7% of the total sample
● Regarding the variable Age_2016 since it provided numerous results (n=2021), in order
to better analyze it, it was recoded into a new variable named AGE_CAT, which
represents the following different age categories, additionally, the results regarding the
frequencies of these in the survey will be provided:
○ Gen Z - younger than 25 years old: 182 respondents, corresponding to 9% of

the sample.
○ Millennials - 26 to 40 years old: 582 respondents, corresponding to 28,8% of
the sample
○ Generation X - 41 to 55 years old: 602 respondents, corresponding to 29,8% of
the sample
○ Baby Boomers - 56 to 74 years old: 504 respondents, corresponding to 24,9%
of the sample
○ Silent Generation - Older than 75 years old: 151 respondents, corresponding
to 7,5% of the sample.
○ In this case the result for both Median and Mode was 3, corresponding to
Generation X, which is the category in which a bigger amount of respondents
belong regarding their age.
2. Do a T-test and compare the means between men and women in the sample.
3. Explain the average age of both groups, their upper and lower bounds and why
there is (no) statically significant difference between them.
T-Test & Mean Comparison Report:
● The individuals surveyed for the social survey Austria 2016 (n=2021), have an average
age of approximately 48 years old (47.74), with a std. Dev: 16.9, however a more
detailed look at the data shows that men (n=936) have an average age of 48 years old
(std. Dev.: 16.4) whereas women (n=1085) have an average age of 47.51 years old
(std. Dev.:17.47). In conclusion we can appreciate both groups in the sample have pretty
much the same mean or average age, which highlights the lack of a significant difference
between them (Sig. 0,512 > 0,05), which is appreciated in the T-test
Q: Are there significant differences between the counties of Austria and the mean income
levels reported?
Use: BLD_2016; Net_income_individual.

Complete the following task:
1. Describe the variables on a univariate level in a sensible(!) way! You may use tables and
recode the income variable if necessary.
● In the first variable BLD_2016, regarding to the

state in which the respondents reside, there was a total
sample of n=2021 respondents, in which the larger
percentage of 20,5% chose Vienna as their state of
residence, with n=414 of the participants. The following
two states which were the most popular were Lower
Austria, with n=373 respondents as 18,5% and Upper
Austria, with n=352 representing 17,4%. The lowest
amount of respondents were the ones living in
Burgenland with only n=67 (3,3%) and Vorarlberg n=82
(4.1%).
● Turning now to the second variable

Net_income_individual, in order to ease the
analysis process the variable was recoded into a
new one, in which 4 categories were formed, the
new variable New_net_ind divides the individual
net income of the participants into the following
categories
○ Low income: from €1 to €1000
○ Medium Low income: from €1001 to €2000
○ Medium High income: from €2001 to €3000
○ High income: from €3001 to €4000
● From the results obtained in this univariate analysis we can observe that in the total
sample (n=1336) the great majority of the respondents said to have an individual income
within €1001 to €2000, with n=791 answers, corresponding to 58,2%, additionally, the
second most popular answer was of the “Lower income” category, with n=394
respondents, corresponding to the 29% of the sample. In order to obtain the Mean,
Median and Mode for this specific variable, the univariate analysis was done once again
with the original variable Net_income_individual, which delivered the following results:
○ Mean: €1411,43
○ Median: €1341
○ Mode: €1168
2. Use a one way anova and ad post hoc tests to find out if there are statistically sig.
differences. Only report those that are significant!
● The following analysis tests if there is a statistically significant difference in the individual
income between people who live in different States of Austria. The total social survey
Austria (n=1336) shows the following:
○ While those who live in Tyrol (n=92) have the highest mean income of €1563
(std. Dev.: 640.3), those who live in Styria (n=200) have the lowest mean income
of €1337 (std. Dev.: 678).
○ According to the LSD test the following results were statistically significant on a
p<0,05 level (all of them are based on the mean difference and significance of
the Reference group (state) vs. the comparison group (state)
■ Vienna, in comparison with:
● Tyrol: -168.825, Sig: 0,021
■ Burgenland, in comparison with:
● Styria: 187.891, Sig: 0,047
■ Styria, in comparison with:
● Burgenland: 187.891, Sig: 0,047
● Tyrol: -225.443, Sig: 0,004
■ Upper Austria, in comparison with:
● Tyrol: -159.420, Sig: 0,037
■ Tyrol, in comparison with:
● Vienna: -168.825, Sig: 0,021
● Styria: -225.443, Sig: 0,004
● Upper Austria:-159.420, Sig: 0,037
○ Since the previously shown results were the only ones with a significant
difference we can observe the main difference existing when comparing Tyrol
(the highest earning state) to the rest, when compared to Styria (the lowest
earning state) the difference in the means is of €-225.443, with a statistically
significant difference of 0,004
Part 2: U-Test and H-Test:
Q: Do women value leisure time less than men?

Compare the importance of leisure time between the sexes. Use a U-Test to do so!
1. Describe the variables on a univariate level in a sensible(!) way! You may use tables.
● Regarding the first variable, as it was described in

the first task, the variable SEX_2016 has a total of
n=2021 respondents, 46,3% (n=936) of them being
male and 53,7% (n=1085) of them female.
● Turning now to the second variable df76_4, this

specific variable tackles the question regarding the
importance of leisure time for each individual,
measuring it on a scale from 1 (not important) to 7
(very important), in this case a great majority of 45%
(n=907) considered leisure time as very important,
with only 0,8% (n=16) considering it not important.
These results are given out of a total sample of
n=2016 respondents
2. Use a U-Test to find out if there are statistically sig. Differences. Use: SEX_2016;
df76_4.
● The results show that women (n=1081) are less likely to consider leisure time as highly
important (mean rank: 992.17 ) than men (n=935, mean rank: 1027.38). However, the
difference between the groups is not significant (0.151 > 0.05). This means that the
surveyed men put a slightly higher value on leisure time than women, however, it cannot
be generalized.
Q: Does the marital status influence the value of leisure time?

Compare the different marital status groups in regards to the importance of leisure time.
1. Describe the variables on a univariate level in a sensible(!) way! You may use tables.
● In the first variable Cf38 the question

answered by the respondents was regarding
their marital status, as it can be observed in
the table, the majority of the respondents
42,7% (n=861) indicated they were married,
followed by those who indicated their status
as single 22.9% (n=462), the lowest
frequency can be observed in the individuals
who are in a registered civil partnership, with
only 0.9% (n=19)
● Turning again to the second variable df76_4,

this specific variable tackles the question regarding
the importance of leisure time for each individual,
measuring it on a scale from 1 (not important) to 7
(very important), in this case a great majority of 45%
(n=907) considered leisure time as very important,
with only 0,8% (n=16) considering it not important.
These results are given out of a total sample of
n=2016 respondents
2. Use a H-Test to find out if there are statistically sig. Differences. Use: Cf38; df76_4.
● The results of the H-test show a significant difference (<0,05) and also show that
although there is an overall high consideration on the importance of leisure time, the
individuals who are in a registered partnership (mean rank: 1144) and those who live
with a partner (mean rank: 1118) have the highest rank means, whereas the widowed
(mean rank: 897.89) have the lowest consideration for leisure time as important.

Statistics 1 Assignment 3

Uploaded by

Copyright:

Available Formats

Statistics 1 Assignment 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Statistics 1 Assignment 3

Uploaded by

Copyright:

Available Formats

Salzburg University of Applied Sciences

Master Degree Program "Innovation and Management in Tourism"

Student Name:​ Catalina Martinez

Student numbers: ​1910649025

Submission Date: ​21.01.2020

Course Lecturer: ​Dimitri Prandner

This assignment concerns itself with differentiation testing.

Part 1 T-Test and A Nova:

USE: Age_2016, SEX_2016

1. Provide a univariate description (NO TABLES!) of the two variables.

○ Gen Z - younger than 25 years old: ​182 respondents, corresponding to 9% of

T-Test & Mean Comparison Report:

Use: BLD_2016; Net_income_individual.

● In the first variable ​BLD_2016​, regarding to the

● Turning now to the second variable

Part 2: U-Test and H-Test:

Q: Do women value leisure time less than men?

● Regarding the first variable, as it was described in

● Turning now to the second variable ​df76_4​, this

Q: Does the marital status influence the value of leisure time?

● In the first variable ​Cf38 the question

● Turning again to the second variable ​df76_4​,

You might also like

Student Name: Catalina Martinez

Student numbers: 1910649025

Submission Date: 21.01.2020

Course Lecturer: Dimitri Prandner

○ Gen Z - younger than 25 years old: 182 respondents, corresponding to 9% of

● In the first variable BLD_2016, regarding to the

● Turning now to the second variable df76_4, this

● In the first variable Cf38 the question

● Turning again to the second variable df76_4,