HR Analytic Chi-Square, Linear Regression

Image credit: wordclouds.
com
6. Using SPSS & Excel:

Chi-Square Test,
💻 Linear Regression
© Dr. Monica Tang 2020. All Rights Reserved.
Dr. Monica Tang
1
This content is copyright protected and shall not be shared, uploaded or distributed
Value Chain in an Organization
How does an organization make money?
Strategy Execution
Business Plan Descriptive Prescriptive Financial Outputs
Predictive
Customer Needs Customer Responses
Operations Operational Outputs
Talent Management HR Services

Intervention
Hiring / and Internal
Acquisition Mobility
Network
Analysis
Performance Turnover and

Evaluation Retention
Fitz-enz, J., & Mattox, J. R. (2014). Chapter 1 Where’s the value. In Predictive analytics for human resources.
Hobokjen, New Jersey: Wiley.
© Dr. Monica Tang 2020. All Rights Reserved. 2

Data Types (1)
A categorical variable is one that is made up of

categories. For example, gender has 2 categories:
male and female. Another example, departments in a
company may include HR, IT and Finance. The
Categorical indicator showing whether the applicant is hired or
not hired is another example They can be thought as
labels which carry no arithmetic meaning.
Data Types
A continuous variable can take on any numerical
value on a measurement scale. For example, age
Continuous
can be any value from 0 and likely up to 120.
Another example, can be salary. Usually, we treat
perception scale, like satisfaction rating (1-5 from
very dissatisfied to very satisfied), as a continuous
variable as well.
[Textbook] Chapter 3 Analysis strategies.

Data Types (2)
Examples:
Variable Data Type
Gender Categorical
Department Categorical
Indicator of Job Offered or Not Categorical
Age Continuous
Salary Continuous
Satisfaction Rating Continuous

Statistical Tests by Data Type
Independent Dependent Statistical Test
Variable (IV) Variable (DV)
Categorical Categorical • Chi-square (χ2) test (when only one IV)
Continuous Categorical • Logistic regression (when more than one IV)

• Survival analysis (when the distribution of time between two events are to
be examined)
Categorical Continuous • Independent samples t-test (when only two groups)
• Paired samples t-test (two separate instances of data collection of the
same group)
• One-way independent ANOVA (when three or more groups)
• One-way repeated measures ANOVA (three or more separate instances
of data collection of the same group)
Continuous Continuous • Pearson’s correlation or simple linear regression (only one IV)
• Repeated measures ANOVA (two or more time points for two or more
groups)
• Multiple regression (linear) (two ore more IVs)
Categorical Continuous • Multiple regression (linear)
and Continuous
• Assumed ‘parametric tests’: distributed symmetrically around the centre and is shaped like a bell curve.

Statistical Significance
• The aim of statistical testing is to determine whether the model

results point to a ‘statistically significant’ outcome. i.e.,
• the pattern of results can be relied upon;
• It is robust enough to make assumptions about what is going
on in the population;
• It is not by chance alone.
• If the test did indicate ‘statistical significance’ at 95% confidence

interval, we have grounds to reject the ‘null hypothesis’.
• i.e., There is only a 5% likelihood (0.05 probability) that the
findings might be due to chance alone.
p < 0.05
Using SPSS & Excel:
Chi-Square Test, Linear Regression
• Diversity and Inclusion

• Demonstration: Gender Diversity Mini Case Study
• Exercise 6.1 Descriptive Statistics
• Exercise 6.2 Chi-Square Test
• Demonstration: Customer Reinvestment Mini Case Study
• Exercise 6.3 Multiple Linear Regression
• Key Points
• Reading List

Diversity and Inclusion

Diversity and Inclusion
• Diversity means “the collective mixture of differences and

similarities that includes, for example, individual and
organizational characteristics, values, beliefs, experiences,
backgrounds, preferences and behaviors”.
• Inclusion means “the achievement of a work environment in which

all individuals are treated fairly and respectfully, have equal
access to opportunities and resources, and can contribute fully to
the organization’s success”.
Edwards, M. R., & Edwards, K. (2019). Predictive HR analytics : Mastering the HR metric (Second ed.). London, United Kingdom, New
York, NY, USA, New Dehli, India: Kogan Page Limited.
Why Diversity and Inclusion
• To comply with equality legislation in place in different parts of the

world. Governments require regulated reporting of data from
organizations to show that there is no unfair treatment on the
basis of protected characteristics such as gender, ethnicity, age
and disability.
• To bring benefits to organizations: better understand customers,

draw from a wider candidate pool, develop a more positive
company image, increase employee engagement and improve
retention, innovation and team performance.

Gender Diversity Mini Case Study
How do we know if there are potential diversity issues in the hiring process?

Consistency of Gender Proportions Mini
Case
in Applicant Pool Study
• In a consultancy firm, there are 280 applications for graduate jobs.
• Are there potential diversity issues in the hiring process?

• Let’s get the insight from data.
Edwards, M. R., & Edwards, K. (2019). Predictive HR analytics : Mastering the HR metric (Second ed.). London, United Kingdom, New
York, NY, USA, New Dehli, India: Kogan Page Limited.

Consistency of Gender Proportions
Data Pattern – Using Excel (1)
• First, let’s look at the data pattern by a frequency count of:

• Gender, Shortlisted and Job Offered
• Excel > Open File: “Chapter 8 RECRUITMENT APPLICANTS.xlsx”
Applicant’s
gender: 1:Shortlisted 1: Job offered
1:Male 0: Not shortlisted 0: Job not offered
2: Female
Each row
represents
1 applicant.

• Excel > Insert > PivotChart
1. Select 2. Select the cell

the data you would like the
range. result to show in.

• Excel > Insert > PivotChart
1. Click and
drag the fields
as shown.
3. Click Count.
2. To display
group by
Gender
3. A good title
• The result is shown. draws the 2. Click “+” to add
reader’s attention Axis Titles, Chart
1. Look at the and strengthens Title and Data
Descriptive the impact of your Labels.
Statistics. message.

Data Pattern – Using Excel
• Use Excel Data: “Chapter 8 RECRUITMENT APPLICANTS.xlsx”

• Prepare this PivotChart:
• Then submit your completed Excel on Moodle Class Exercise 6.1 –

Excel (2% of Total Course Grade).

Data Pattern
• From the data pattern, we suspect there is gender bias in our

hiring process.
• We do Chi-square test to analyze the relationships of the

categorical variables, 2 variables at a time.
• Gender and Shortlisted;
• Gender and Job Offered.

Chi-square – Using SPSS (1)
• To do this in SPSS:
• SPSS > Open Data: “Chapter 8 RECRUITMENT APPLICANTS.sav”
• SPSS > Analyze > Descriptive Statistics > Crosstabs
1. Drag “Gender” to
2. Click “Chi-square”
“Row(s) and
and “Continue”.
“ShortlistedNY” to 2. Click “OK”.
“Column(s)”.
Chi-square – Using SPSS (2)
Result:
Chi-sq (Pearson’s) = 14.997
p-value < 0.05
The results suggest that there

may be some sort of bias in the
shortlisting of the applicants in
favor of male applicants. It is
not just by chance.

in Applicant Pool
• Do the male preferences continue to the job offer stage?
• Please run an appropriate statistical test in SPSS.

• Use SPSS Data: “Chapter 8 RECRUITMENT APPLICANTS.sav”
• After running the results in SPSS, please go to Moodle and answer

the questions in Class Exercise 6.2 Q&A (2% of Total Course
Grade).

Customer Reinvestment Prediction
What selection factors predict customer reinvestment?

Q: What are the predictors in customer reinvestment?

We hypothesize that:
H1: Employee’s understanding customer needs positively predict reinvestment.
H2: Employee’s sales confidence positively predict performance.
H3: Employee’s giving a recommendation positively predicts performance.
H4: Employee’s product knowledge positively predict performance.
Hypothesized Model
Understanding Customer Needs +
Sales Confidence + Customer Reinvestment
Giving a Recommendation
+
+
Product Knowledge

Let’s first see the variables.

Customer Reinvestment Predictor
Multiple Linear Regression – Using SPSS (1)
• To test the hypothesized model in SPSS:
• SPSS > Open Data: “CHapter 7 Customer satisfaction only N2507.sav”
• SPSS > Analyze > Regression > Linear
1. Drag INvestMore here as DV.
2. Drag the Sat1-4 and

SexOfSalesperson here as IVs.
3. Click “OK”.
• Let’s check the model fit of our hypothesized model.
Result:
2
R Square = 0.17
Our IVs account for a good
deal of variation (17%) in
customer reinvestment
intentions.
2
Result:
p < 0.001
This model is highly
significant.

• Let’s check look for the key drivers of customer reinvestment
intentions.
1) Identify variables with
p<0.05.
2) Interpret the meaning
(B).
Understanding customer
needs and having a
recommendation are the
two key drivers of customer
reinvestment indices.

Multiple Linear Regression – Using SPSS
• Use SPSS Data: “CHapter 7 Customer satisfaction only

N2507.sav”
• Please run an appropriate statistical test for this model in SPSS.
Hypothesized Model
Understanding Customer Needs +
Sales Confidence + Customer Reinvestment
Giving a Recommendation
+
+
Product Knowledge
• Then submit the SPSS output .spv fiile on Moodle Class Exercise
6.3 – SPSS (2% of Total Course Grade).

Reading List

Thank you.
Any Questions?


HR Analytic Chi-Square, Linear Regression

Uploaded by

Document Information

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

HR Analytic Chi-Square, Linear Regression

Uploaded by

Copyright:

Available Formats

Image credit: wordclouds.

6. Using SPSS & Excel:

Customer Needs Customer Responses

Operations Operational Outputs

Talent Management HR Services

Performance Turnover and

© Dr. Monica Tang 2020. All Rights Reserved. 2

A categorical variable is one that is made up of

© Dr. Monica Tang 2020. All Rights Reserved. 3

Indicator of Job Offered or Not Categorical

Satisfaction Rating Continuous

© Dr. Monica Tang 2020. All Rights Reserved. 4

Continuous Categorical • Logistic regression (when more than one IV)

[Textbook] Chapter 3 Analysis strategies.

© Dr. Monica Tang 2020. All Rights Reserved. 5

• The aim of statistical testing is to determine whether the model

• If the test did indicate ‘statistical significance’ at 95% confidence

• Diversity and Inclusion

© Dr. Monica Tang 2020. All Rights Reserved. 7

© Dr. Monica Tang 2020. All Rights Reserved. 8

• Diversity means “the collective mixture of differences and

• Inclusion means “the achievement of a work environment in which

• To comply with equality legislation in place in different parts of the

• To bring benefits to organizations: better understand customers,

© Dr. Monica Tang 2020. All Rights Reserved. 10

© Dr. Monica Tang 2020. All Rights Reserved. 11

• In a consultancy firm, there are 280 applications for graduate jobs.

• Are there potential diversity issues in the hiring process?

© Dr. Monica Tang 2020. All Rights Reserved. 12

• First, let’s look at the data pattern by a frequency count of:

© Dr. Monica Tang 2020. All Rights Reserved. 13

• Excel > Insert > PivotChart

1. Select 2. Select the cell

© Dr. Monica Tang 2020. All Rights Reserved. 14

• Excel > Insert > PivotChart

© Dr. Monica Tang 2020. All Rights Reserved. 16

• Use Excel Data: “Chapter 8 RECRUITMENT APPLICANTS.xlsx”

• Then submit your completed Excel on Moodle Class Exercise 6.1 –

© Dr. Monica Tang 2020. All Rights Reserved. 17

• From the data pattern, we suspect there is gender bias in our

• We do Chi-square test to analyze the relationships of the

© Dr. Monica Tang 2020. All Rights Reserved. 18

The results suggest that there

© Dr. Monica Tang 2020. All Rights Reserved. 20

• Do the male preferences continue to the job offer stage?

• Please run an appropriate statistical test in SPSS.

• After running the results in SPSS, please go to Moodle and answer

© Dr. Monica Tang 2020. All Rights Reserved. 21

© Dr. Monica Tang 2020. All Rights Reserved. 22

Q: What are the predictors in customer reinvestment?

© Dr. Monica Tang 2020. All Rights Reserved. 23

Let’s first see the variables.

© Dr. Monica Tang 2020. All Rights Reserved. 24

1. Drag INvestMore here as DV.

2. Drag the Sat1-4 and

© Dr. Monica Tang 2020. All Rights Reserved. 26

© Dr. Monica Tang 2020. All Rights Reserved. 27

• Use SPSS Data: “CHapter 7 Customer satisfaction only

© Dr. Monica Tang 2020. All Rights Reserved. 28

[Textbook] Chapter 3 Analysis strategies.

© Dr. Monica Tang 2020. All Rights Reserved. 29

© Dr. Monica Tang 2020. All Rights Reserved. 30

You might also like