Running head: PARAMETERS
1
Parameters
Layal Mansour
Walden University
PARAMETERS
2
Parameters
Research topic
Assessment of an individual weight in respect to age, class and status of work as a public health
initiative towards understanding the risk of health complications related to body weight in a bid
to promote healthy living.
Research Questions
What is the association between an individual weight with respect to age, class of work and work
status?
Study Population
In carrying out a research, a study population entails a collection of individual forming the
primary focus for a scientific question and posits various binding characteristics (Taylor,
Bogdan, & DeVault, 2015). Since, the study entails a public health issue affecting the entire
population; incorporate the entire population because the primary objective for carrying out the
public health-related research is to understand the risk of health complications related to
individual weight based on age, class of work and the status of work.
Study Variables
There are a number of variables involved in the study aimed at providing a clear
understanding on the public health issue related in body weight.
Dependent element: This a variable in a research that depends on other attributes often denoting
the point on a researcher’s interest. In our study, the dependent variable is the person’s weight.
Independents variable: This comprises of study elements that influence the dependent attribute
in a research. In the study, a person's age, class, and status of work are the independent element
believed to have a significant influence on the dependent element body weight.
PARAMETERS
3
Rationale for Selecting the Data variables in relation to Research Question
The research question aims at evaluating the relationship between an individual body
weight based on a person's age, class and status of work. This would assist in developing public
health insights related to risks of health complication related to the weight of a person in a bid to
establish a basis for effective measures that would assist in educating the public regarding the
benefit of weight management based on age, class, and status of work. In public health body,
weight has been associated with increasing cases of type 2 diabetes, cardiovascular ailment,
hypertension, and strokes among other health complication. This has resulted in various public
health department carrying out assessments aimed at providing informed solutions towards
minimizing the increasing health complications. As a result, the selected variables are highly
suitable for carrying out the assessment and provide educative insight on patterns of weight
across the three independent elements. Besides, the variables are suitable in the provision of vital
association of weight with age, class, and status of work in a bid to provide ways that would
assist in educating the population about the risk of excess weight on their health.
Sample Selection
A sample in research encompasses a group of individuals selected randomly from the
entire population that posits the same characteristics as the study question (Flick, 2015). Besides,
a sample should representative to heighten the ability to generalize the finding from the study
sample to the entire population. The study will utilize available information collected for a
different reason referred to as secondary data. This is because the study area is vast and requires
many resources to accomplish the objective of the study related to the public health initiative.
Specifically, the study will utilize a number of variables derived from dataset collected in Puerto
PARAMETERS
4
Rico PR in the United States, a data collected in a personal record census by the American
government.
Appropriate Sample Size at 95 percent Confidence
The expected marginal error for the study is 0.05 with a standard deviation of 3 individual.
Therefore,
Sample size n =1.962*sd2/ E2
Sample size n= 1.962*32/0.052
Appropriate sample size n = (3.8416*9)/0.0025
Sample size n = 13,830
Power Analysis
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Source
Type III Sum of
df
Mean Square
F
Sig.
Squares
Intercept
34112778.607
1
34112778.607
Error
12010575.726
11185
1073.811
31767.955
Noncent.
Observed
Parameter
Powera
.000
31767.955
a. Computed using alpha = .05
Based on the table above, the observed power is 1.000 implying that we have a very high chance
that the analysis will reveal statistical significance at 95 percent confidence level.
1.000
PARAMETERS
5
Reference List
Flick, U. (2015). Introducing research methodology: A beginner's guide to doing a research
project. Sage.
Taylor, S. J., Bogdan, R., & DeVault, M. (2015). Introduction to qualitative research methods: A
guidebook and resource. John Wiley & Sons.
Running head: DATA ANALYSIS PLAN
Data Analysis Plan
Layal Mansour
Walden University
1
DATA ANALYSIS PLAN
2
Data Analysis Plan
In performing a research work, data analysis plan is vital because it enlightens a reader in
the form of analysis that would explore a hypothesis in the course of answering a research
question. Phillips and Stawarski (2008) notes that data analysis plan outlines various aspect of a
data evaluation in a research, which includes the data cleaning, analysis assumptions and any
form of transformation that a researcher performs.
Research Questions
The study will examine the following research questions related to housing in Puerto Rico/ PR.
o Is the number of bedrooms per housing unit dependent on the lot size?
o Is there a relationship between the house heating fuel and the number of bedrooms per
housing unit?
Selected Variables
The variable used to answer the research questions includes the bedrooms in a housing
unit, lot size, and house heating fuel.
Variable Description
Bedrooms: This element describes the number of bedrooms a housing unit has categorized into
five attributes, which are no bedroom, one bedroom, two bedrooms, three bedrooms, four
bedrooms, five or more bedrooms,
DATA ANALYSIS PLAN
3
Lot size: The variable entails the size of land a house lies grouped as a house on less than oneacre land, house on one to less than 10 acres of land as well as a house on ten or more acres of
land.
House heating fuel: The variable encompasses the type fuel used in heating in a housing unit
classified in various fuel attributes, such as utility gas, bottled, tank, or LP Gas, Electricity, Fuel
oil, kerosene, etc., Coal or coke, wood, solar energy, other fuel or no fuel used in heating.
Types of variables
The selected variables for the analysis process depict both independent as well as
dependent characteristics. In the first research question that examines whether the number of
bedrooms per housing unit dependent on the lot size, the attribute bedrooms will be the
dependent variable while the lot size will be the independent attribute. Moreover, in the second
research question, the number of bedrooms will be the independent variable and the housing
heating fuel will be the dependent element, implying that the type of house heating fuel unit
depends on the number of bedrooms per housing unit. Notably, a dependent variable denotes the
point of interest for a research often influenced by the independent attribute.
Level of Measurement
Different levels of measurement describe different statistical variables, which include the
ordinal, interval, nominal and ratio scale. In this case, the selected elements depict nominal
scales used in assigning events into discrete categories.
Statistical Analysis Plan
Data analysis entails the process of converting raw data into simple and easy to read
statistical information that one can interpret and make meaningful and significant conclusions
(Clinical tools, Inc., n.d). Moreover, statistical evaluation assists in making imperative
DATA ANALYSIS PLAN
recommendation used in effecting necessary changes to enhance the attainment of the objective
for carrying out a research.
4
DATA ANALYSIS PLAN
5
Descriptive Method
Cooper and Schindler note that researchers use descriptive statistics to describe features
of information utilized in a study (2014). Descriptive methods provide summaries about a
selected sample often giving the measure of central tendency, dispersion of the sample data
among other statistics such as the frequencies depending on the level of measurement. As a
result, since the selected data depict a nominal level of measurement, the descriptive method will
include frequencies and percentages of the attributes on a selected variable.
Analytic Methods for Answering the Research Questions
Statistical Test
The research will use chi-square independence test in the process of examining the
hypothetical statement aimed at answering the research questions. Field (2013) defines Chisquare independence test as a statistical test utilized in the determination of whether there exists
a significant association between two nominal attributes.
Reason why the Test is Appropriate
The study assumes that the variables are dependent on one other, for instance, the number
of bedrooms dependent on the lot size while house heating fuel depending on the number of
bedrooms per a housing unit. This makes the selected statistical test the most appropriate in
answering the research questions. Besides, a chi-square test for independence is the most
appropriate to test the variable because the selected data satisfy the categorical level of
measurement, which is a condition necessary to carry out the statistical test.
How the Statistical Test help in Answering the Research Question
The results from the statistical test will assist in testing hypothetical statement formulated
from the research question with an aim to reject the null hypothesis at α=0.05. In case the
DATA ANALYSIS PLAN
6
findings lead to rejecting the null hypothesis, this will imply that the test accepts the alternative
hypothesis. Accepting the alternative hypothesis depict that the research questions are true.
Method for Presenting the Findings
The statistical findings both the descriptive and inferential results will be presented using
standalone tables and histogram. Notably, standalone implies that the tables and histogram will
offer all necessary information, such that an audience of the research would easily read and
understand the presented content without necessary reviewing the provided inference.
DATA ANALYSIS PLAN
7
Reference List
Clinical tools, Inc. (n.d). Guidelines for Responsible Data Management in Scientific Research.
Cooper, D. R., & Schindler, P. S. (2014). Business research methods (Vol. 12). New York:
McGraw-Hill Irwin.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. sage.
Phillips, P. P., & Stawarski, C. A. (2008). Data Collection : Planning for and Collecting All
Types of Data. San Francisco: Pfeiffer. Retrieved from
http://search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=218523&site=ehostlive
Running head: DATA MANIPULATION
Manipulating the Dataset in SPSS
Layal Mansour
Walden University
1
DATA MANIPULATION
2
Manipulating the Dataset in SPSS
Manipulating data involve the process of converting available data into a more organized
and easy to understand in a bid to make informed decisions. As a result, the end-results from a
data manipulation is displaying of information to the users, in a meaningful and usable way.
Research Questions
The research questions in the study are
o Is the number of bedrooms per housing unit dependent on the size of land a house
occupies?
o Is there a relationship between the fuel used in heating the house and the number of
bedrooms per housing unit?
Null Hypothesis
o The number of bedrooms per housing unit is not dependent on the size of the land the
house occupies.
o There is no association between the fuel used in heating a house and the number of
bedrooms per a housing unit.
Variables Definition and Categories
Bedrooms in a housing unit, lot size, and house heating fuel are the variable that will
answer the defined research questions. The variable bedrooms contain a number of categories
included no bedroom, one bedroom, two bedrooms, three bedrooms, and four bedrooms, five or
more bedrooms. Similarly, the element lot size has a number of categories, which include a
house on less than one-acre land, house on one to less than 10 acres of land as well as a house
on ten or more acres of land. House heating fuel as well contain elements such as utility gas,
DATA MANIPULATION
3
bottled, tank, or LP Gas, Electricity, Fuel oil, kerosene, etc., Coal or coke, wood, solar energy,
other fuel or no fuel used in heating.
Data Conversion from Continuous to Categorical
The selected elements for the analysis are categorical in nature, revealing a nominal level
of measurement. Therefore, none of the elements requires converting from continuous to a
categorical variable.
Exploration of the Categorical Variables Results
The data exploration entails both the descriptive and analytical analysis in SPSS and
presentation of the findings in standalone tables for easier interpretation of the results.
Descriptive Statistics
1. Lot size
The table below illustrates the frequencies for the variable Lot size involving the number
of acres.
Table 1
Lot size
Frequency
N/A (not a one-family house or
mobile home)
Valid Percent
Cumulative Percent
2207
15.3
15.3
15.3
11298
78.4
78.4
93.7
House on one to less than ten
acres
770
5.3
5.3
99.1
House on ten or more acres
132
.9
.9
100.0
14407
100.0
100.0
House on less than one acre
Valid
Percent
Total
Inference
The frequency table 1 above reveals various the number and percentage for each category in the
variable lot size. From frequency table, 15 % of the sampled houses were mobile home/ not a
one family house, while 78.4% entailed houses built on less than one acre of land.
DATA MANIPULATION
4
Correspondingly, houses on one to less than ten acres and house on ten or more acres depicts
5.3% and 0.9% respectively.
2. Bedrooms
The table below reveals the frequencies for the variable bedroom, indicating the number
of bedrooms per unit house.
Table 2
Bedrooms
Frequency
Percent
Valid
Valid Percent
Cumulative Percent
No bedrooms
392
2.7
2.7
2.7
1 Bedroom
884
6.1
6.1
8.9
2 Bedrooms
2972
20.6
20.6
29.5
3 Bedrooms
7676
53.3
53.3
82.8
4 Bedrooms
2103
14.6
14.6
97.4
380
2.6
2.6
100.0
14407
100.0
100.0
5 or more Bedrooms
Total
Inference
Table 2 reveals the frequencies for the number of bedrooms per a unit house based on the
house record dataset collected from Puerto Rico. From the table, 3 bedroomed houses revealed
the highest percent of the sample houses at 53.3%, followed closely by 2 bedroom houses at 20.6
percent of the total houses in the house record dataset. Subsequently, houses with no bedrooms
depicted 2.7 percent, while 5 or more bedrooms house revealed 2.6 % of the collected house
record data in Puerto Rico/ PR. Nonetheless, 1 and 4 bedroomed houses in the dataset posited
6.1% and 14.6 % respectively from the total house record data in the dataset.
3. House heating fuel
DATA MANIPULATION
5
The table below illustrates the frequencies for the variable house heating fuel indicating
various forms of fuel energy used in heating the houses based on the sampled house record data
from Puerto Rico/PR.
Table 3
House heating fuel
Frequency
Percent
N/A (Vacant)
Utility gas
“Bottled, tank, or LP gas”
Electricity
“Fuel oil, kerosene, etc.”
Valid
Coal or coke
Wood
Solar energy
Other fuel
No fuel used
Total
Valid Percent
1547
22
10.7
.2
10.7
.2
Cumulative
Percent
10.7
10.9
357
2.5
2.5
13.4
2457
17.1
17.1
30.4
7
.0
.0
30.5
1
1
115
25
9875
14407
.0
.0
.8
.2
68.5
100.0
.0
.0
.8
.2
68.5
100.0
30.5
30.5
31.3
31.5
100.0
Inference
From the frequency table 3 above, the largest number of houses did not use any form of fuel in
heating the house at 68.5 % of the sample houses in Puerto Rico/ PR house record dataset.
Nonetheless, 17.1 percent of the houses used electricity to heat the house, with 2.5 and 0.2% of
the home using Bottled, tank, or LP gas and Utility gas respectively and the house heating fuel.
Moreover, 0.8% and 0.2 of the houses used solar energy and other fuel respectively as house
heating fuel while 10.7% of the houses in the database were vacant.
Analytical Statistics
The analytical process involved chi-square tests for independence aimed at answering the
research questions. The table below illustrates the chi-square test between the number of
bedrooms per house and the size of the land the house occupies.
DATA MANIPULATION
6
Table 4
Chi-Square Tests
Value
df
Asymp. Sig. (2-sided)
Pearson Chi-Square
1725.580a
15
.000
Likelihood Ratio
1446.661
15
.000
Linear-by-Linear Association
749.667
1
N of Valid Cases
14407
a. 2 cells (8.3%) have expected count less than 5. The minimum expected count is 3.48.
.000
Inference
From table 4 above, the Pearson Chi-square value is 1725.580 at 15 degrees of freedom.
Besides, the p-value is 0.000 less than α= 0.05, implying that we reject the null hypothesis.
Therefore, we are 95 % confident that the number of bedrooms per housing unit dependents on
the size of the land the house occupies.
Moreover, the table below illustrates the chi-square test between the fuel used in heating
the house and the number of bedrooms per housing unit.
Table 5
Chi-Square Tests
Value
df
Asymp. Sig. (2-sided)
Pearson Chi-Square
381.670a
45
.000
Likelihood Ratio
354.759
45
.000
Linear-by-Linear Association
175.228
1
.000
N of Valid Cases
14407
a. 29 cells (48.3%) have expected count less than 5. The minimum expected count is .03.
Inference
From the chi-square table 5 above, the Pearson Chi-square value is 381.670 with 45
degrees of freedom. The p-value from the table is 0.000 less than α= 0.05 depicting that we reject
DATA MANIPULATION
the null hypothesis. Thus, we are 95 percent confident that the fuel used in heating the house is
associated with the number of bedrooms in a house.
7
Running head: INTERPRETING RESULTS
Interpreting Results
Layal Mansour
Walden University
1
INTERPRETING RESULTS
2
Interpreting Results
The resulting interpretation will justify the findings from statistical assessment aimed at
addressing hypothetical statements developed from research questions in a bid to provide
explanations for the research. Bryman states that a hypothesis is a supposition developed with
minimal evidence, for further statistical investigation (2016). The interpretation of the result will
include both descriptive and inferential statistics obtained from SPSS output. The research
questions under investigation are
Is the number of bedrooms per housing unit dependent on the size of land a house
occupies?
Is there a relationship between the fuel used in heating the house and the number of
bedrooms per housing unit?
Null Hypothesis
o The number of bedrooms per housing unit is not dependent on the size of the land the
house occupies.
o There is no association between the fuel used in heating a house and the number of
bedrooms per a housing unit.
Interpretation of Descriptive Results
Descriptive statistics describe the basic features revealed by data under investigation,
through the provision of summaries about a selected sample (Bryman & Bell, 2015). In this case,
the selected samples include three variables, the number of bedrooms, size of land and house
heating fuel. Since the variables depict nominal levels of measurement, the descriptive statistic
INTERPRETING RESULTS
3
for the elements encompasses the attributes frequencies mainly the counts as well as the
percentage.
Lot size
The first variable under descriptive investigation is lot size depicting the size of land in
acres that a sampled house covers. The table below reveals the frequencies for the element lot
size.
Table 1
Lot size
Frequency
N/A (not a one-family house or
mobile home)
Valid Percent
Cumulative Percent
2207
15.3
15.3
15.3
11298
78.4
78.4
93.7
House on one to less than ten
acres
770
5.3
5.3
99.1
House on ten or more acres
132
.9
.9
100.0
14407
100.0
100.0
House on less than one acre
Valid
Percent
Total
Interpretation
From the table, it is evident that the largest proportion of the sampled houses covered less
than one acre of land with 11,298 houses out of the total sampled house of 14407 house in Puerto
Rico. This presented 78.4 percent of the total sampled houses in the region. Therefore, from a
statistical point of view, we may conclude that the largest number of houses in Puerto Rico
covered less than one acre of land with only a few houses occupying a larger space of more than
one acre. However, a significant number of houses occupied more than 10 acres of lands totaling
to132 houses out of the total sampled houses.
Bedrooms
The selected sample variable reveals the number of bedrooms each sampled house
contained. The frequency table below indicates the number of bedrooms per unit house.
INTERPRETING RESULTS
4
Table 2
Bedrooms
Frequency
Percent
Valid
Valid Percent
Cumulative Percent
No bedrooms
392
2.7
2.7
2.7
1 Bedroom
884
6.1
6.1
8.9
2 Bedrooms
2972
20.6
20.6
29.5
3 Bedrooms
7676
53.3
53.3
82.8
4 Bedrooms
2103
14.6
14.6
97.4
380
2.6
2.6
100.0
14407
100.0
100.0
5 or more Bedrooms
Total
Interpretation
Form the table above, it is evident that a majority number of sample houses contained
three bedrooms with 7676 houses out of the total the 14,407 sampled units. This is more than 50
% of the sampled data which indicated that there is a likelihood that majority of the houses built
in Puerto Rico contain three bedrooms. However, the distribution of the number of bedrooms
reveals a normal curve implies that the number of bedrooms contained in the house in Puerto
Rico is normally distributed (see Appendix chart 1).
1. House heating fuel
The variable house heating fuel indicates various forms of fuel energy used in heating the
houses based on the sampled house record information from Puerto Rico/PR region. The table
below depicts the frequencies for the variable house heating fuel.
INTERPRETING RESULTS
5
Table 3
House heating fuel
Frequency
Percent
N/A (Vacant)
Utility gas
“Bottled, tank, or LP gas”
Electricity
“Fuel oil, kerosene, etc.”
Valid
Coal or coke
Wood
Solar energy
Other fuel
No fuel used
Total
Valid Percent
1547
22
10.7
.2
10.7
.2
Cumulative
Percent
10.7
10.9
357
2.5
2.5
13.4
2457
17.1
17.1
30.4
7
.0
.0
30.5
1
1
115
25
9875
14407
.0
.0
.8
.2
68.5
100.0
.0
.0
.8
.2
68.5
100.0
30.5
30.5
31.3
31.5
100.0
Interpretation
From the frequency table, most of the houses did not utilize any form of fuel in heating totaling
to 9875 houses out of the 14,407 houses sample from the Puerto Rico region. This indicates that
the region has favorable climatic conditions that do not require heating the house to keep warm.
However, some areas on higher altitudes experience cool weather requiring heating with
electricity revealing the commonly used heating fuel in the area with 17.1 percent of the total
sampled houses.
Interpretation of the Inferential Statistics
Sekaran and Bougie define inferential statistics are data analysis result obtained for
statistical tests aimed at providing solutions to a question under scrutiny. Inferential assessments
enhance the ability of a researcher to make conclusions that answers a research question using
statistical evaluation (2016). The inferential statistics involved Chi-square tests aimed at
rejecting the null hypotheses developed from the research question for further statistical
investigation.
INTERPRETING RESULTS
6
Research Question 1
Is the number of bedrooms per housing unit dependent on the size of land a house
occupies?
The table below illustrates the chi-square test between the number of bedrooms per house
and the size of the land the house occupies.
Table 4
Chi-Square Tests
Value
df
Asymp. Sig. (2-sided)
Pearson Chi-Square
1725.580a
15
.000
Likelihood Ratio
1446.661
15
.000
Linear-by-Linear Association
749.667
1
N of Valid Cases
14407
a. 2 cells (8.3%) have expected count less than 5. The minimum expected count is 3.48.
.000
Interpretation
From the chi-square table 4 above, we reject the null hypothesis because the p-value for
Pearson Chi-square is less than 0.05 at 95% level of confidence. This implies that the number of
bedroom per housing unit depends on the size of land a house occupies. Therefore, houses built
on large tracts of land are likely to have a higher number of bedroom compared to houses
occupies small portions of land say less than a one-acre piece in Puerto Rico.
Research Question 2
Is there a relationship between the fuel used in heating the house and the number of
bedrooms per housing unit?
The table below depicts a chi-square test between the variables fuel used in heating the house
and the number of bedrooms per housing unit.
INTERPRETING RESULTS
7
Table 5
Chi-Square Tests
Value
df
Asymp. Sig. (2-sided)
Pearson Chi-Square
381.670a
45
.000
Likelihood Ratio
354.759
45
.000
Linear-by-Linear Association
175.228
1
.000
N of Valid Cases
14407
a. 29 cells (48.3%) have expected count less than 5. The minimum expected count is .03.
Interpretation
The table reveals a p-value of 0.00 less than 0.05, implying that we reject the null hypothesis.
This is an indicator that there is a relationship between the fuel used in heating the house and the
number of bedrooms per housing unit. Therefore, we can deduce houses with higher the number
of bedrooms could be using electricity as house heating fuel due to efficiency in the heating
process.
INTERPRETING RESULTS
8
Reference list
Bryman, A. (2016). Social research methods. Oxford university press.
Bryman, A., & Bell, E. (2015). Business research methods. Oxford University Press, USA.
Sekaran, U., & Bougie, R. (2016). Research methods for business: A skill building approach.
John Wiley & Sons.
INTERPRETING RESULTS
9
Appendix
Appendix Chart 1
Appendix Chart 2
INTERPRETING RESULTS
Appendix Chart 3
10
Purchase answer to see full
attachment