HUMAN RESOURCE MANAGEMENT 3A

Unit 4: Problem Identification and Preparing Data

Kgomotso Mopalami
Defining problems
So many problems, so little time…

• Identifying problems accurately is a critical step to ensure that HR measurement is sound and effective
• If not to solve problems, why should one analyse metrics?
• Once a people-related problem is identified, associated HR metrics need to be identified and tracked
• Thus, the first step is to unpack and extract the people-related problem in a broader organisational context.
• Whenever a business problem is discussed, try to identify what impact the people can have on it or a solution
that is being developed.
• What needs to be put in place?
• What is already in place?
• In what ways can the people management function support the attempts to solve the problem?
• In fact, problems are everywhere, and most of what we do at work is about resolving them.
• The first steps in the process of solving problems is to recognise that there is a problem, and to identify,
define or construct it.
• Identifying the problem can assist in focusing on the right set of matters – not on the associated emotion.

3
What is a problem?

a dilemma with no an undesirable a question that you

obvious way out situation without a cannot answer
solution currently.

Balanced
opposition
leading to conflict

4
 • The person’s or group’s state of
Initial state knowledge at the start of a problem

• The goal that the person or group wishes

Goal state to achieve

Components of a
problem Actions • The operations that the problem solver can
use to get to the goal state

• Features of the environment that can

Task either directly or indirectly constrain or
environment suggest different ways of solving a
problem

5
Types of problems

Well-defined problems have a definite initial state and the goal

state, and the actions that can be taken to reach the goals are
known.

Ill-defined problems are those problems in which the actions, the

goal state or sometimes even the initial state of the situation is
unknown.

6
Information characteristics of a problem

Lack of A problem can occur if information is missing. This could lead to incomplete
communication or the existence of unknowns due to which the problem cannot be
information resolved.

Wrong Inaccurate information is another possible characteristic of a problem, as the wrong

information information could lead to misunderstandings and poor decision making.

Differing Differing information may be the most prevalent characteristic of a problem. This can
lead to differing viewpoints, changed impressions and balanced dilemmas, leading
information to persistence and hidden emotions.

7
Process of solving a problem

Defining the problem Outcome of problem

identification
Problem recognition Problem statement
Problem labelling Question
Problem-cause analysis Hypothesis

8
Problem statement
Establishes concisely what is currently happening, describing
what the concern at hand is.

The problem statement can be about:

• an area of concern
• a condition to be improved upon
• a difficulty to be eliminated
• a troubling question that exists in theory or in practice.

In the form of a sentence, or few, providing context and

identifying the gap between the current state and the desired
state.

Purposes of the problem statement

• To identify the problem

• To explain the problem

Helps those involved to understand the problem clearly.

9
Getting to the problem statement
Developing a problem statement
• Identify the gap (or pain) that exists.
• Ensure that the timeframe, location and trend of the problem are included.
• Quantify the impact or gap that exists in terms of cost, time, quality, resources, etc.
• Determine the importance and urgency of the problem to the various stakeholders involved.
• Focus on the end state – the problem statement should focus on what we are aiming for and not how we will get there.
Consider the 5Ws (and 1H) of every problem.
• What is the problem that needs to be solved?
• Why is this a problem? (highlight the pain)
• When is the problem being experienced? (first observation, frequency)
• Who is experiencing the problem? (stakeholders)
• Where is the problem observed? (location, department, activities)
• How is the problem observed? (symptoms, magnitude, trends)

Challenges in developing problem statement

• Symptoms
• Solutions
• Causes
• Blame

10
Question

While a problem statement clarifies the problem at hand, a question can give clarity about what
aspects of the problem need to be investigated and the process that must be followed to answer the
question.
It is what we do not know about the problem that may evolve into becoming a question.

The data you collect is only as good as the questions you ask about them

A poorly identified research question can result in inaccurate data collection, misinterpretation or
faulty interpretation of the collected data, which can then cause additional problems

The more specific and focused your question is, the more actionable the output of the data analysis
process will be.

11
Characteristics of a question
The question should Type of Examples
highlight a genuine The question could be question
dilemma, area of open- or closed-ended.
ambiguity or point of Either way, it must provide Describing or What are the steps in the recruitment process?
How has the induction process changed over time?
confusion about the issue direction. exploring What are the main characteristics of the performance
at hand.
management process?
How do employees experience the employer’s branding
activities?
The question must be How do managers deal with poor performance?
The question should have
guiding, not prescriptive. Explaining or What is the relationship between gender and
associated data. If not, it
In other words, the engagement?
should be feasible to
solution should not be testing What is the role of the manager in retaining talented
collect associated data.
hidden in the question. employees?
What is the impact of the induction process on new
employee retention?
The question should be What are the causes of presenteeism in the
The question should not organisation?
simple, focusing on only
ask for a subjective value What are the advantages and disadvantages of the
one or two concepts at a Evaluating
judgement, so should not
time. The more concepts, current compensation practices?
contain words such as and acting
the more confusing the How effective are the organisation’s wellness initiatives?
“good”, “bad”, “better”, etc.
analysis process will be. What are the most effective strategies to improve
retention of critical employees?

12
Hypothesis

A hypothesis is a possible outcome of the problem identification process

Described as a statement linking various aspects of the problem, especially if one can predict how these aspects link to each other.

A hypothesis usually tests the relationship between two or more constructs or aspects of the problem.

It states a prediction – a tentative answer to a question that has not been tested yet

A hypothesis can be proven right or wrong – so one must be prepared for both outcomes.

Characteristics of a good hypothesis

• The language used in the hypothesis must be clear and focused.
• The relationship between the hypothesis and the problem at hand must be clear.
• The hypothesis must be testable – one should be able to collect data points associated with the hypothesis and apply statistical tests to it.
• The hypothesis must include independent and dependent variables.
• The variables in the hypothesis must be manipulatable subject to ethical standards.

13
 Developing a hypothesis

01 02 03 04 05
Ask a Do some Formulate Clarify the Write a null
question preliminary the hypothesis hypothesis
about the research or hypothesis
issue at hand analysis

14
Agreement

IF YOU FEEL THERE YOUR AWARENESS IDENTIFY A FIND THE OVERLAP IDENTIFY THE ASK OPEN-ENDED CONTINUALLY
IS A PROBLEM, OF THE PROBLEM REFERENCE POINT IN VIEWPOINTS. FALSEHOODS AND QUESTIONS WORK ON A
THEN THERE MAY BE DIFFERENT THAT IS ACCEPTED ANALYSE THE UNKNOWNS THAT WITHOUT STRONG AND OPEN
DEFINITELY IS A FROM OTHERS’. ASK AND OBSERVABLE. INFORMATION MAY BE CAUSING THREATENING OR WORKING
PROBLEM. DO NOT HOW MUCH OTHERS USE FACTS, RECEIVED FROM THE PROBLEM TO ACCUSING. ASK RELATIONSHIP. THIS
IGNORE IT, BUT KNOW OF THE SPECIFIC EVENTS ALL INVOLVED AND PERSIST. KEEP ABOUT OTHERS’ WILL PREVENT
PROBE FOR THE PROBLEM OR HOW AND DISCUSSION IDENTIFY WHAT IS COMMUNICATING FEELINGS, DEFENSIVENESS
OTHERS’ THINGS ARE POINTS. DO NOT COMMON. UNTIL YOU FIND THOUGHTS AND AND INTIMIDATION.
AWARENESS OF DIFFERENT FROM USE VALUE THEM. OBSERVATIONS,
THE PROBLEM. HOW THEY SHOULD JUDGEMENTS. AVOIDING LEADING
BE. OR JUDGEMENTAL
QUESTIONS.

15
Making decisions about the solutions

Optional solutions – list all alternative strategies that may resolve the
problem and address the root cause.

Decision making – choose the best solution by objectively evaluating

the options identified in the previous step.

Action planning – identify and organise steps taking into consideration

the task/activity, person responsible, time and resources required, and
outcomes expected in order to implement the decision.

16
Identifying problems
Tools for identifying and solving problems

Scientific method Problem space – task Steps in the scientific method

environment
Initial state Hypotheses – clear goal state Observation
Goal state Experiments – unclear goal state Research
Set of actions Hypothesis
Test
Data analysis
Conclusion

18
Tools for identifying and solving problems

Brainstorming Benchmarking Experiment

Generation of ideas and different Systematic process of measuring Piloting
framing of problems within a group one’s performance against
Tests whether a problem has been
recognised leaders
Consensus and integrating ideas identified correctly and whether the
are important Determining best practices that can solution addresses the problem
lead to superior performance when comprehensively before spending
All individuals should be given an adapted and utilised
opportunity to explain how they resources and efforts
Comparison enables you to identify
framed the problem
shortcomings, differences in
Creative, flexible, encouraging and processes, etc.
inclusive May also guide you in relation to
solving the problem

19
Tools for identifying and
solving problems
• Setting SMART goals
• S – Specific
• M – Measurable
• A – Achievable
• R – Relevant
• T – Time bound

20
Considerations of problem identification

Persons involved in problem Ethics of problem Communication Timing

identification identification
Who should be involved? Organisational vision and values Poor communication causes barriers to Bad timing can cause emotional flare-
Who must help in identifying the solutions – that can in itself become a ups and improper communication,
problem? problem distracting from the problem
Who must approve the problem Seek, respect and accept other
statement? viewpoints

21
Value of problem identification

Experts spend a lot of time in the problem identification stage.

Problem identification usually involves information regarding the goals

of the problem-solving effort, the process used to arrive at the goals,
and any constraints and restrictions placed on the solution

Actively engaging in the process of problem identification results in

increased creativity in determining the solution

22
Role of the HR professional in problem
solving
• Identifying the right problem
• Facilitating the problem identification process
• Knowing and being aware of what data has been and can be collected in
relation to the problem
• Playing the role of people expert

23
DATA EXPLORATION AND COLLECTION

• Step 2 of the HR measurement cycle

 • Data collection is the process of gathering and
measuring data on specific targeted aspects of a
system, activities or individuals.
• Purpose of data collection
• To ensure that there are sufficient data points that
can enable data analysis which in turn, enables the

What is data analyst to answer questions or evaluate various

outcomes, depending on what the problem at hand

collection? is.
• One must consider others’ perspectives and data that
may be already available to you.
• Process of data collection
• Should be technically sound
• Should be of sound quality
• Enables evidence-based decision making

25
Features of data collection
26
Sources and examples of HR data

27
Sources of process-related information
28
Locating data

Data is collected for a specific business need that

Intentional has been identified

One comes across data that has already been

Opportune collected but for a purpose different from the
business need

Look for opportune data before Seems wasteful and inefficient to collect data that
is already available to the HR function.
collecting data for the problem at Employees may get frustrated if the same
hand question is being asked repeatedly to them

29
Locating data

Why was the data collected previously?

Existing data in HR systems How was the data collected?
Is the data relevant?

Collecting data from non-HR systems

Survey Questionnaire Interview

Group discussion Observation Experiment
Collecting new data Introspective and retrospective tasks
Self-report instruments

30
Extent of data required

Population Small population which is easily Large population

accessible
Entire group of employees that you want to do Collect data from the entire population Sample – group or subset of interest from
the analysis of and reach a conclusion about within the population
A sample can be used to represent the whole
population if:
• The sample satisfies all the different
variations present in the population and
meets the selection criteria which must be
well-defined.
• The sample is unbiased in every way
possible.
• The sample is randomly chosen.
Sampling - process of collecting data from the
sample and then using to generalise over the
entire population

31
Types of HR data

Categorical – data that can be divided into groups or

categories, no intrinsic order
Quantitative data Continuous – data that can be divided and reduced to
finer levels

About types and may be represented by a name,

symbol, or a number code.
While quantitative data may be the answers to ‘how
Qualitative data many’ or ‘how much’ questions, qualitative data may
be the answers to more open-ended questions.
Qualitative data may be subjective in nature.

32
 Once-off
Time span
of data
collection
Continuous How often?

33
Outcome Name Employee Gender Date of Function Role

of data
Number Birth
Angela 001 M 06.07.1989 Sales Manager

collection
Bongi 073 F 15.10.1993 Finance Officer
Clement 034 M 08.02.1981 Marketing Officer
Dumisani 003 M 30.07.1968 HR Director

34
Considerations when collecting data – Before data collection
Spend some time to consider what analysis will be done with the data and what the results of such analysis will be

Determine how much data you need for good findings. Identify the population, sampling and data saturation

Consider secondary data available before going for primary data – This reduces cost and ensures rapid analysis.

Consider the ethical aspects associated with each data point that is being collected

Evaluate if the data collection process can cause harm to participants

Think critically about the data that must be collected

Consider the cost associated with every additional data point

Use identifiers wisely

Consider how each data point will be obtained

Pilot any data collection process

Be respectful of participants’ time when collecting data.

35
Considerations when collecting data – During data collection

Be transparent about the data collection process

Ensure confidentiality and anonymity.

Communicate regularly with the participants and potential participants

Check that you have enough responses to do the necessary analysis

Ensure that the collected responses cannot be modified or altered

36
Considerations when collecting data – After data collection

It is equally important to inform

Keep the respondents posted the respondents as well as
about progress other stakeholders about the
results of the analysis.

37
Data cleaning and preparing
 Data cleaning Data preparing
Finding and removing errors in the data Readying data for analysis
Errors creep in during the data collection Converting variables to analysis-ready
process formats
If not identified, problems may arise when Removing rows with missing information
analysing the data Outliers – those cases that are very different
Even if no problems arise, the analysis will from the rest of the data – may also be
be incorrect, leading to wrong decisions considered for removal

39
 Errors that must be cleaned

Type Example
Values that may be impossible or An employee entered birth year as 1895, instead of 1995
incorrect

Cases that should not be included Someone who did not attend a training course responded to the training evaluation.
in the analysis

Duplicate cases An employee filled a survey online and then also filled up the physical form.

Missing data An employee did not complete the survey and submitted the survey prematurely, or an
employee who went through the entire survey, but did not complete all items.

Incorrect data entry An employee typed ‘Mael’ instead of ‘Male’

40
 Original dataset

Employee Date of Training Training

Name Number Gender Birth Function Role attended evaluation

Angela 001 F 06.07.1989 Sales Manager Yes 4

Bongi 073 F 15.10.1993 Finance Officer Ni 2

Clement 034 M 08.02.1981 Marketing Officer Yes 4

Angela 001 M 06.07.1989 Sales Manager Yes 3

Dumisani 003 M 30.07.1968 HR Director Yes 3

41
Cleaned dataset

Employee Date of Training Training

Name Number Gender Birth Function Role attended evaluation

Angela 001 F 06.07.1989 Sales Manager Yes 4

Bongi 073 F 15.10.1993 Finance Officer Yes 2

Clement 034 M 08.02.1981 Marketing Officer Yes 4

Dumisani 003 M 30.07.1968 HR Director Yes 3

42
Prepared dataset
Gender Code Attendanc Code
Training e
Employee Date of Training
Gender Function Role evaluatio Female 1 Yes 1
Number Birth attended
n Male 2 No 0

Function Code Training Code

001 2 06.07.1989 2 2 1 4
evaluation

Finance 1 Highly 4
satisfied
073 1 15.10.1993 1 3 1 2
Sales 2 Satisfied 3
Marketing 3 Unsatisfie 2

034 2 08.02.1981 3 3 1 4 d

HR 4 Highly 1
unsatisfied
003 2 30.07.1968 4 1 1 3 Did not 0
attend

43
 Description Excel Function
Excel TRIM function takes the cell reference (or text) as the
input. It removes leading and trailing spaces as well as the TRIM(text)

Using
additional spaces between words (except single spaces).
Converts all text into Lower Case LOWER(text)
Converts all text into Upper Case UPPER(text)
Selects all blank cells in your dataset.

Excel to
If you want to enter 0 or Not Available in all these cells, just Ctl+A (Select all)  F5 (Go to) Special… (bottom left)

type it and press Control + Enter (remember if you press Blank  OK

only enter, the value is inserted only in the active cell).

clean and
In any blank cell, type 1 Select the cell where you typed
1, and press Control + C Select the cell/range which you
want to convert to numbers Select Paste –> Paste
Converts numbers stored as text into numbers
Special (Key Board Shortcut – Alt + E + S) In the Paste

prepare
Special Dialogue box, select Multiply (in operations
category) OK
Home –> Conditional Formatting –> Highlight Cells Rules –
Highlights duplicates
> Duplicate Values.

data
Go to Data –> Remove Duplicates. If your data has
headers, ensure that the checkbox at the top right is
Delete duplicates
checked. Select the Column(s) from which you want to
remove duplicates and click OK.
Searches for a value in the first column of a table array and
returns a value in the same row from another column in the VLOOKUP(value, table, column no, exact match)
table array.
Parses long text bits into multiple cells. Data –> Text to Column (select Delimiter)

44
Thank you!

45