Techniques, Tools and Tactics -

Methods of Study and Research

E D I T E D N OT E S B Y M S . L O H WA N L I N A N D M S .
G E R A L D I N E PA N G I R A S
❖ American software engineers provide assistance to a colleague only when they expect to
need that colleague’s help at some future time; Indian software engineers provide help to
whoever needs it without any expectation of future reciprocation (Perlow & Weeks, 2002).
❖ United States workers are less concerned about job insecurity than are Chinese workers
(Probst & Lawler, 2006).
❖ Asians have a very different understanding of time than Americans (Brislin & Kim, 2003)
❖ In considering who gets a bonus and how much, Chinese managers make decisions based
more on the personal needs of the individual, whereas in the U.S. these decisions are
based more on the performance of that individual (Zhou & Martocchio,2001)
❖ Compared to American managers, Japanese managers are much more likely to solve a
strategic problem by being cooperative and making sure that individuals share equally in
rewards (Wade-Benzoni, Okumura, Brett, Moore, Tenbrunsel, & Bazerman, 2002)
 Answering questions and making decisions
• Structure interview vs unstructured interview
( Schmidt & Hunter, 1998)

You will probably have to deal with the findings,

Why Study advertisements that containing references to
research supporting a particular products

Research
Methods? To use their advice you must understand how they
studied the problems and arrived at their conclusions

Common sense is often wrong ( Kruger, Wirtz, &

Miller, 2005)
 Objective observation
• Base conclusions on objective
evidence of the present situation

Requirements • Control
• Possible influences on outcomes
of Research should be identified and measured

• Duplication and Replication of

verification earlier studies
 Not all behaviour can be studied

• Complex and dangerous research

Observing behaviour can change it

• Respondents are often aware they are being a part of research

Research The Hawthorne effect

Limitations
• Employee’s behaviour changed just because something new has
been introduced into the workplace

Artificial settings

• stimulated job environment

College students as subjects

• Limit the generalisability of the research findings

Steps in IDEA @ QUESTION
HYPOTHESIS OR THEORY OR

Research PREDICTION EXPLANATION

• Decide what to research -

Employee referrals will 1.Realistic job preview
“I wonder....”
result in employees who 2.Differential
• Form hypothesis from What employee stay with the company recruitment
theory recruitment source is best? longer than will the 3.source
other recruitment 4.Personality similarity
methods 5.Socialization theory

What will happen Why It will happen

 Idea Hypothesis Theory

Does all this High levels of Noise causes a

noise affect my noise will increase distraction
employees’ the number of making it difficult
performance? errors made in
assembling to concentrate
electronic
components
 • Literature review

•Get ideas on what method,

results had been done
Steps in
• Decide method to be
Research used
•Experimental
•Observation
•Survey
 Experiments

Research Naturalistic observations

Methods

Questionnaires, Polls and surveys

The Experimental Method

The scientific way to Independent

determine the effect Dependent variable
variable
or influence of a
variable on the
subjects performance
or behaviour The stimulus variable
The resulting behaviour
of the subjects, which
that is manipulated to
depends on the
determine its effect on
manipulation of the
the subjects behaviour
independent variable
Designing and Experiment:
Effects of Lighting on Productivity
❖Independent variable
❖Lighting
❖Dependent variable
❖Production rate
❖Element of control
❖Experimental group – exposed to the independent variable
❖Control group – not exposed to the independent variable, but similar
in every other way
 ❖Control group and experimental group must
be as similar as possible
❖Random group design
❖Subjects assigned at random to
experimental and control groups
Selecting ❖Matched group design
Participants ❖Subjects in one group are matched with
subjects in the other group on
characteristics that could affect their
performance
❖Desirable, but costly and time consuming
 Subjects -208 employees in lingerie factory
Dependent variables - Turnover and productivity
Independent variable - Level of training
◦ Group 1: 1-day of training (control)
Effects of ◦ Group 2: 2-days in training facility

Training on ◦ Group 3: 3-days in training facility

◦ Group 4: 3-days mixed training (facility & OJT)
Turnover & Results [graph]

Productivity ◦ Turnover was lowest in Group 3, but so was

productivity
◦ Turnover was almost as low in Group 4, and
productivity was nearly as high as in the control
group 3-days mixed training (facility & OJT)
 Observation of behaviour in natural setting without
experimental manipulation of independent variable

Avoids artificiality – observed behaviours are typical

of what naturally happens

Naturalistic However, without manipulations of IVs, it is difficult

to conclude what brought about change
Observation
The observation cannot be repeated

Observer effects
 ❖Participants
❖1,319 clerks in 576 convenience stores

Naturalistic ❖ Observations were made on levels of

courtesy, which were then related to sales
Observation ❖ Findings
of Store ❖the higher the incidence of courteous
Clerks behaviours, the lower the sales
❖• the busier the store, the less time the
clerk had to exhibit courteous behaviour
 Survey research method relies on
interviews, behavioural observations,
and questionnaires to sample what
Questionnaires, people say
Surveys and
Opinion Polls
Focus is not on what people do, but
what they say they do
 Difficult to measure precisely our personal opinions and attitudes

The Some people deliberately lie

Problems
Socially or politically correct Give an opinion on the subject People more likely to be
response even if they don’t have one honest when in private

With Difficult to find people to respond

Surveys
Phrasing of the question often influences the response
 Personal
interviews

Ways To Paper and

Pencil
Collect Questionnaires

Survey Web-based •Most frequently used approach to employee polling

•More reliable than telephone surveys

Data surveys •Subject to “ballot stuffing”

Telephone •Problems arising from increased cell phone use

surveys
 Open-end

• Respondents state their views in their own words

• Provides riches quality of data
Basic • Difficult to analyze

Types of Fixed-alternative

Survey • Respondents must limit their answers to the

choices presented
Questions • Simplifies the process
• May not completely reflect respondents range of
opinion - limits responses

It’s a good idea to pretest the questions

on a small sample
 Will the participant understand the question?
Will the question itself change the way a
person thinks?
Do the response options cover the construct?
What are we going to do with the data?
Question ◦ What question are we trying to answer?
Considerations ◦ How much time, effort, and money are we
willing to spend in coding and analyzing
responses?
Does the format increase or decrease the
probability of responding?
 ❖Probability sampling
❖Each person in the population has a
known probability of being included in the
Methods for sample
Constructing A ❖Quota sampling
Representative ❖Because the sample must reflect the
Sample proportions of the larger population,
quotas are established for categories such
as age, gender, and ethnic origin
 Advantages

• Can be conducted 24/7

• Faster, less costly

Web- • Availability of larger, more diverse pool of

participants

Based • Direct data entry reduces coding

Research Disadvantages

• Impossible to assess honesty on demographics

• Range restriction – more educated, affluent
people have computer access
• Lower response and higher drop-out rates
 Descriptive statistics

•Ways of describing research data in a

concise, meaningful manner

Methods Of inferential statistics

Data •Used to compare performance levels of

Analysis the experimental group and the control
group

Meta-analysis

•The large scale reanalysis and averaging

of the results of previous research studies
 Graphs

Measures of central tendency

Presentation
•Mean
of •Median
Descriptive •Mode
Statistics
Measures of variability

•Standard deviation
•Correlation coefficient
 Mean
•Arithmetic average

Measures of Median
Central •The score at the midpoint of a
Tendency statistical distribution
Mode
•The most frequently obtained
score in the distribution of data
 Standard Deviation (SD)

•It is a precise distance along the

distribution’s baseline
•Using the SD, we can determine the
percentage of scores that fall above or
Measures of below any particular raw score
Variability
• Correlation

•The relationship between two variables

•strength of relationship
•direction of relationship
 Normal Distribution
• Bell-shaped with most scores falling
toward the middle with few at high and
low extremes
Sample • Mean, median and mode are equal
Distributions Skewed Distribution
• Asymmetrical with more scores falling
closer to high or low extremes
• Median is most useful measure of
central tendency because it is less
affected by extreme scores
Methods for analysing data that express
relationships (e.g., differences between
groups) in terms of probabilities

Statistical significance
The level of confidence we
have in the results of an
Inferential
Statistics
experiment based on
probability values

The idea that a difference

between the means of
experimental and control
groups could have occurred by
Probability chance
If p=.01, a difference would
occur by chance only 1 time
out of 100.
 The large scale reanalysis of the results
of previous studies

The average effect size of multiple

Meta studies weighted by sample size
Analysis
Helps to determine overall trends

Used to reach more objective conclusions

where a large body of research exists
DISCUSSION ON RESULTS CONCLUSION
Ethics of Research

Research
Informed
Debriefing Review
consent
Boards
 Ethically required

Informed
Consent
Can be waived when
Waiver will not Research could not
Research involves
adversely affect be done without the
minimal risk
rights of participants waiver
 •procedure that is conducted in psychological research
with human participants after an experiment or study has
been concluded.

•involves structured or semi-structured verbal

conversation between the researcher and the

Debriefing participants - to inform participants of the true nature of

the study
• participants are given an opportunity to ask questions
• participants informed about what the hypothesis for the
experiment
• If the participants has any misconceptions about the study,
the researcher takes reasonable steps to correct those
misconceptions
 •Also known as ethics review board or
institutional review board/committee
•reviewing the methods proposed
for research to ensure that they are ethical
•formally designated to approve (or reject),
Research Review monitor, and
Boards/Committees review biomedical and behavioral research
involving humans.
•assure that appropriate steps are taken to
protect the rights and welfare of humans
participating as subjects in a research study.