0% found this document useful (0 votes)
37 views16 pages

BUS 4055 Week 5

Uploaded by

kbrennan2833
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)
37 views16 pages

BUS 4055 Week 5

Uploaded by

kbrennan2833
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/ 16

BUS 4055 Week 5: Quantitative Data

Analysis
Kevin Brennan, M.Mgt.

Photo by Claudio Schwarz on Unsplash

Introduction to Quantitative
Data
– Objective: Understand how quantita‐
tive data is processed and analyzed to
turn raw data into meaningful informa‐
tion.
– Key Point: Raw data is often mean‐
ingless without processing.
Photo by Mika Baumeister on Unsplash
– Methods: Quantitative techniques like
graphs, charts, and statistics help de‐
scribe and explore relationships in data.
– Discussion: Discuss how quantitative data is typically used to make decisions in
business research.
– Example: Student satisfaction survey results, sales data, or website traffic analy‐
sis.

Importance of Quantitative
Data in Business Research
– Business Application: Quantitative
data provides the backbone for decision-
making in business, from budgeting to
customer satisfaction surveys.
– Common Examples: Frequencies,
test scores, prices, rental costs, or stu‐
Photo by Victoriano Izquierdo on Unsplash
dent feedback.
– Key Focus: Identifying trends, testing
hypotheses, and making data-driven decisions.
– Discussion: How businesses use quantitative data to drive decisions in fields
like finance, marketing, and operations.
– Example: Evaluating a company's customer service based on the number of re‐
solved support tickets or customer ratings.
Overview of Data Types
– Key Data Types:
– Categorical: Non-numerical data used
for classification (e.g., gender, product
categories).
– Numerical: Data that can be counted
Source: https://stackoverflow.com/ques‐
or measured (e.g., age, income).
tions/34095777/handling-unassigned-null-values-of-cat‐
– Discussion: Explain the difference
egorical-features-in-regression-machine
between categorical and numerical data
and when each is used.
– Example: Survey responses about satisfaction levels using categories like "Very
Satisfied" or numerical ratings from 1 to 5.

Categorical Data
– Definition: Data used to classify ob‐
servations into distinct categories with‐
out numerical meaning.
– Key Properties: No intrinsic numerical
relationship (e.g., car models, survey
responses).
– Discussion: Categorical data is help‐
ful in classifying observations into
Baijayanta Roy, https://towardsdata‐ groups.
science.com/all-about-categorical-variable- – Example: Analyzing survey results to
encoding-305f3361fd02 see the number of students who use
specific college resources, such as the
library or counselling services.
Nominal Data
– Definition: A categorical data type where categories are named but have no in‐
trinsic order.
– Key Features: No ranking or order among categories (e.g., gender, departments,
or colors).
– Discussion: Nominal data provides labels to classify variables without assigning
them any order or value.
– Example: Classifying survey respondents by their department (e.g., Engineering,
Business, Arts).
Ordinal Data
– Definition: Categorical data where the categories have a meaningful order or
rank but uneven intervals.
– Key Properties: Data points can be ranked, but differences between points are
not equal.
– Discussion: Ordinal data is useful for ranking but cannot precisely measure the
difference between ranks.
– Example: Customer satisfaction ratings: "Very Satisfied," "Satisfied," and "Unsat‐
isfied," where each category reflects increasing satisfaction.
Numerical Data: Discrete vs. Continuous
– Discrete Data: Countable, often finite
(e.g., number of students in a class).
– Continuous Data: Measurable, can
take any value within a range (e.g.,
weight, distance).
– Key Focus: Discrete data represent
Source: https://betterexplained.com/articles/under‐
standing-discrete-vs-continuous-growth/ whole numbers, while continuous data
can represent fractions.
– Example: Discrete - Number of products sold; Continuous - Sales revenue over
time.
– Discussion: Which type of data would be most useful in measuring student suc‐
cess?

Interval Data
– Definition: Numerical data where the
intervals between values are meaningful,
but there is no true zero.
– Key Properties: Can add or subtract
values, but ratios are meaningless (e.g.,
20°C is not twice as warm as 10°C).
– Use: Typically used when differences
between values are meaningful, but
comparisons using ratios are not.
– Example: Temperature, SAT scores.
– Discussion: What situations would
require using interval data, and what are
Source: http://reliawiki.com/index.php/Interval_‐
its limitations?
Data_Example

Ratio Data
– Definition: Numerical data that includes a true zero point, allowing for meaning‐
ful ratios.
– Key Properties: Ratios are meaningful (e.g., someone earning $100K earns
twice as much as someone earning $50K).
– Use: Ratios can be compared directly because of the presence of a true zero
point.
– Example: Income, age, weight.
– Discussion: Why is ratio data important when comparing quantitative variables
like income or population?
Source: https://careerfoundry.‐
com/en/blog/data-analytics/what-is-ratio-
data/

Data Collection Methods


– Surveys: Widely used in quantitative
research to gather large amounts of
data.
– Questionnaires: Typically used to
capture structured responses.
– Discussion: How choosing the right
data collection method affects the type
and quality of data collected.
– Example: A student satisfaction sur‐
vey using a Likert scale to quantify their
Photo by Scott Graham on Unsplash
responses about services like tutoring or
library access.

Preparing Data for Analysis


– Data Cleaning: Removing errors, du‐
plicates, and inconsistencies in data.
– Data Coding: Converting categorical
responses into numerical values for eas‐
Data Preparation Data Analytics

ier analysis.
Raw Data Prepared Data Insights – Importance: Clean data leads to accu‐
rate results and reduces bias.
– Example: Cleaning a student survey
dataset by standardizing responses to
ensure consistency before analysis.
– Discussion: Why is data cleaning an
Source: https://devopedia.org/data-prepa‐ essential step in the research process?
ration Coding Categorical Data

Data Matrix: A table where rows represent cases (individual responses) and col‐
umns represent variables.

Data Organization: Each row in the matrix represents a respondent, and each col‐
umn a survey question.

Use: A data matrix organizes data in a structured way, making it easier to perform
analysis.

Example: A matrix displaying survey responses, with rows for individual students
and columns for satisfaction ratings.

Discussion: How does organizing your data in a matrix improve data analysis effi‐
ciency?

Categorical Coding: Assigning numerical values to categorical data (e.g., 1 =
Male, 2 = Female).

Codebook: A reference guide that explains each numerical code and its meaning.

Importance: Coding allows easier manipulation and analysis of non-numerical
data.

Example: Coding student satisfaction responses on a scale of 1-5, with 1 meaning
"Very Dissatisfied" and 5 "Very Satisfied."

Discussion: What are the potential pitfalls of incorrect data coding?
Data Visualization
– Purpose: Graphs and charts make
data easier to understand and communi‐
cate to stakeholders.
– Common Graph Types:
– Bar Chart: Compare categorical data
(e.g., number of students per major).
– Line Graph: Track changes over time
(e.g., enrollment rates).
– Scatterplot: Show relationships be‐
tween variables (e.g., hours studied vs.
GPA).
– Example: A line graph showing the trend of student satisfaction over multiple se‐
mesters.
– Discussion: Why is visualization important in data analysis? How do different
types of graphs help convey specific types of information?

Organizing Data: Frequency Tables


– Definition: Frequency tables display
the number of occurrences of each cate‐
gory or value in a dataset.
– Purpose: Helps summarize data and
identify patterns or trends.
– Use: Frequency tables help visualize
and simplify categorical data.
– Example: A frequency table showing
Source: https://edu.gcfglobal.org/en/sta‐ how many students rated their satisfac‐
tistics-basic-concepts/frequency-tables/1/ tion level as 1, 2, 3, etc.
– Discussion: How does a frequency
table help understand the distribution of categorical responses?

Bar Charts
– Definition: Bar charts visually repre‐
sent categorical data by using bars to
show the frequency of each category.
– Purpose: Makes it easy to compare
the number of occurrences in different
categories.
– Use: Bar charts are particularly useful
for showing differences between groups
or categories.
– Example: A bar chart displaying the
number of students from each major
(e.g., Business, Engineering, Arts).
– Discussion: Why are bar charts effec‐
tive for comparing categorical data in
© Ladislav Kubačka business analysis?

Histograms
– Definition: Histograms represent the distribution of continuous data by using ad‐
jacent bars to show frequencies in intervals.
– Purpose: Useful for understanding the shape of data distributions (normal,
skewed, etc.).
– Use: Typically used to show how frequently different ranges of values occur in a
dataset.
– Example: A histogram showing the distribution of student ages in a survey sam‐
ple.
– Discussion: How does the shape of a histogram help in understanding data pat‐
terns?
av danielpenfield - eget arbete, cc by-sa 3.0,
https://commons.wikimedia.org/w/index.php?
curid=9401898

Pie Charts
– Definition: Pie charts divide a circle
into slices to represent proportions of a
whole.
– Purpose: Best used to show the rela‐
tive sizes of categories.
– Use: Common for visualizing categori‐
cal data where percentages or propor‐
tions are relevant.
– Example: A pie chart showing the per‐
centage of students using different cam‐
pus services.
– Discussion: When is a pie chart more
effective than a bar chart, and what limi‐
tations might it have?

https://infogram.com/blog/choose-the-right-
chart/

Measures of Central Tendency: Mean


– Definition: The mean (or average) is the sum of all data points divided by the
number of points.
– Use: Provides a "typical" value for a dataset, especially useful for numerical data.
– Key Point: Sensitive to outliers; ex‐
treme values can skew the mean.
– Example: Calculating the average sat‐
isfaction score from student feedback,
where scores range from 1 to 5.
– Discussion: When is the mean a reli‐
able indicator of central tendency, and

source: https://faculty.elgin.edu/dkernler/sta‐when might it be misleading?


tistics/ch03/3-1.html

Measures of Central Tendency: Median


– Definition: The median is the middle
value when data points are arranged in
order.
– Use: Provides a central point that is
less affected by outliers than the mean.
– Key Point: Useful for skewed distribu‐
tions where the mean might not be rep‐
resentative.
– Example: Median income levels in a
survey of student financial aid recipients.
– Discussion: How does the median
help provide a clearer picture when the
data includes extreme values?
https://opentextbc.ca/mediastudies101/chapter/political-
economies-of-digital-media/

Measures of Central Tendency: Mode


– Definition: The mode is the most frequent value in a dataset.
– Use: Helps identify the most common value, especially useful for categorical
data.
– Example: Finding the most common satisfaction rating on a student course feed‐
back form.
– Discussion: In what scenarios is the mode more informative than the mean or
median?
https://faculty.elgin.edu/dkernler/statistics/ch03/3-
1.html

Graphical Representation of Central Tendency


– Purpose: Visualize key summary sta‐
tistics such as mean, median, and mode.
– Graph Types:
– Box Plot: Shows the distribution, high‐
lighting the median and quartiles.
– Line Chart: Displays changes over
time.
– Example: A box plot showing the dis‐
https://ledidi.com/academy/measures-of-
tribution of test scores among students
central-tendency-mean-median-and-mode
in different departments.
– Discussion: How does visualizing
measures of central tendency aid in understanding data? Discuss the value of us‐
ing box plots versus line charts.

Measures of Dispersion: Range


– Definition: The range is the difference between the highest and lowest values in
a dataset.
– Use: Provides a basic measure of variability within a dataset.
– Key Point: The range is limited as it only considers the two extreme values.
– Example: The range of student GPAs in a sample from 2.0 to 4.0.
– Discussion: What are the limitations of using range as the sole measure of dis‐
persion?
Source: https://sixsigmadsi.com/range-
measure-of-dispersion/

Measures of Dispersion: Standard Deviation


– Definition: Standard deviation mea‐
sures the average distance of each data
point from the mean.
– Use: Helps understand the spread of
data, with a small standard deviation in‐
dicating that data points are close to the
mean.
– Key Point: Provides a more nuanced
view of variability compared to the
source: https://opentextbc.ca/mediastudies101/chap‐
range.
ter/political-economies-of-digital-media/
– Example: Calculating the standard de‐
viation of test scores in a class to see
how much variation exists in student performance.
– Discussion: Why is standard deviation preferred over range for understanding
data variability?

Graphical Representation of Dispersion


– Purpose: Use graphs to show the
spread or variability of data.
– Graph Types:
– Histogram: Visualizes the frequency
distribution of continuous data.
– Box Plot: Highlights the range, quar‐
tiles, and potential outliers.
– Example: A histogram showing the
distribution of student ages across the
college.
Source: http://derekogle.com/Book107/Univ‐ – Discussion: Why is it important to vi‐
EDA.html sualize data variability? How do graphs
like histograms provide insights into pat‐
terns not easily seen in raw data?

Multivariate Analysis
– Definition: Multivariate analysis in‐
volves examining more than two vari‐
ables simultaneously to understand the
relationships between them.
– Use: Common in business research to
explore complex interactions between
variables.
– Example: Examining how age, in‐
http://www.sthda.com/english/articles/32-r-
come, and education level affect student
graphics-essentials/129-visualizing-multi‐
satisfaction with college services.
variate-categorical-data/
– Discussion: What are the benefits
and challenges of using multivariate
analysis in business research?

Statistical Significance
– Definition: Statistical significance as‐
sesses whether results from data are
meaningful or could have occurred by
chance.
– Key Concept: The p-value is com‐
monly used to determine significance
(e.g., p < 0.05 means the result is statis‐
tically significant).
– Use: Ensures that the findings of a
Source: https://researcher.life/blog/arti‐ study are not random and can be gener‐
cle/what-is-p-value-calculation-statistical- alized to the population.
significance/ – Example: Testing whether differences
in satisfaction between two groups of
students (e.g., online vs. in-person) are statistically significant.
– Discussion: Why is it important to test for statistical significance in business re‐
search?

Chi-Square Test
– Definition: A chi-square test is used to determine whether there is a significant
association between two categorical variables.
– Use: Commonly applied in surveys and observational studies.
– Example: Testing whether student satisfaction is related to their year of study
(e.g., freshman, sophomore, junior).
– Key Concept: The test compares observed frequencies with expected frequen‐
cies under the assumption that there is no association.
– Discussion: What types of research
questions can the chi-square test help
answer?

By Mikael Häggström - File:Chi-square distri‐


butionCDF.png, Public Domain, https://com‐
mons.wikimedia.org/w/index.php?
curid=10633630

T-Test

Title
By Cornell User 2 - Own work, CC0,
https://commons.wikimedia.org/w/index.php?
curid=140537163
– Definition: A t-test compares the means of two groups to determine if there is a
statistically significant difference between them.
– Use: Commonly used to compare two groups in experiments or surveys.
– Example: Comparing the average test scores of students who received tutoring
versus those who did not.
– Key Concept: A low p-value (< 0.05) suggests the difference between the two
groups is statistically significant.
– Discussion: When should a t-test be used, and what are its limitations?

Analysis of Variance (ANOVA)


– Definition: ANOVA tests whether there are statistically significant differences be‐
tween the means of three or more groups.
– Use: Useful for comparing multiple groups or conditions in experiments.

Example: Comparing the effectiveness
of different teaching methods across
multiple classes.
Key Concept: If the p-value from ANOVA
is low, it suggests that at least one group
mean is significantly different from the
others.
– Discussion: When is ANOVA more
appropriate than a t-test?

Identifying Trends in Data


– Purpose: Look for patterns or trends
that can provide insights into business
processes or customer behaviour.
– Methods:
– Line Graphs: Show trends over time.
– Scatterplots: Identify relationships be‐
tween variables.
– Example: Identifying a trend in student
participation rates in campus events
over multiple semesters.
– Discussion: Why is it important to
Source: https://pressbooks.senecacollege.ca/operationsmanagement/chap‐
identify trends in business data? Discuss
ter/forecasting/
how trends can inform strategic deci‐
sion-making.

Correlation Analysis
– Definition: Correlation measures the
strength and direction of the relationship
between two continuous variables.
– Use: Helpful in identifying relationships
but does not imply causality.
– Example: Investigating whether
there’s a correlation between the number
of hours students study and their GPA.
– Key Concept: A positive correlation
indicates both variables increase togeth‐
source: https://statistics.laerd.com/statisti‐
er, while a negative correlation means
cal-guides/pearson-correlation-coefficient-
one decreases as the other increases.
statistical-guide.php
– Discussion: How can correlation
analysis help identify trends, and what
are its limitations?
Regression Analysis
– Definition: Regression analysis exam‐
ines the relationship between a depen‐
dent variable and one or more indepen‐
dent variables.
– Use: Common in predictive modeling,
allowing researchers to forecast out‐
comes based on known variables.
– Example: Predicting student success
based on variables like hours studied,
attendance, and prior academic perfor‐
mance.
– Key Concept: Multiple regression al‐
lows for the inclusion of several independent variables to predict the outcome.
– Discussion: Why is regression analysis valuable for making predictions in busi‐
ness research, and what are the assumptions?

Week 5: Quantitative Data Analysis - Expanded Content


(Slides 31-45)

Common Pitfalls in Quantitative Data Analysis


– Key Pitfalls:
– Misinterpreting Correlations: Re‐
member, correlation does not imply cau‐
sation.
– Sampling Bias: Ensure that your sam‐
ple accurately represents the population.
– Overfitting: Avoid too many variables
in regression models, which can lead to
overfitting.
– Example: Assuming that increased
hours of study directly cause higher
GPAs without considering other factors.
Source: https://www.frontiersin.org/articles/10.3389/fda‐
– Discussion: How can you avoid com‐
ta.2019.00013/full
mon pitfalls in your own research? Dis‐
cuss strategies for ensuring data accura‐
cy and integrity.

Ethical Considerations in Quantitative Research


– Key Principles:
– Data Privacy: Protect the anonymity
and confidentiality of participants.
– Informed Consent: Ensure partici‐
pants understand how their data will be
used.
– Data Integrity: Avoid manipulating
data to fit preconceived hypotheses.
– Example: Ensuring that student feed‐
back data is anonymized before analy‐
sis.
Photo by Isaac Smith on Unsplash
– Discussion: What ethical dilemmas
might arise in quantitative research, and
how can they be mitigated? Discuss potential consequences of unethical data
practices.

Key Takeaways from Week 5


– Recap: Review the most important
concepts covered in the week, including
data types, statistical tests, and visual‐
izations.
– Applications: Discuss how these con‐
cepts will apply to the final project and
real-world research scenarios.
– Activity: Have students list out the
methods they plan to use in their own
research projects.
Photo by Aldin Nasrun on Unsplash
– Discussion: What areas do you feel
most confident about, and what areas
need more clarification?

Next Steps for Research Project Preparation


– Action Items:
– Refine Your Research Questions: Ensure they are clear and measurable.
– Start Data Collection: Begin gathering and organizing your data.
– Preliminary Analysis: Perform initial tests to explore your data.
– Discussion: What is your timeline for completing the analysis phase of your
project? What challenges do you anticipate?
Photo by Kier in Sight Archives on Un‐
splash

You might also like