Social Science Research Methods

presenting and discussing theories, interpreting research results, and studying the effectiveness of
various interventions used by I/O practitioners.

Research methodology is a set of procedures that allow us to investigate the hows and whys of human
behavior and to predict when certain behavior will and will not occur.

will study the basic social science research methods used by I/O psychologists to study work behavior.

Research methodology

statistical analysis is nothing more or less than procedures for testing the repeated objective
observations that a researcher has collected.

Goals of Social Science Research Methods

For example, an I/O psychologist might attempt to satisfy the first goal by describing the production
levels of a company, the rates of employee absenteeism and turnover, and the number and type of
interactions between supervisors and workers for the purpose of arriving at a more accurate picture of
the organization under study.

The goal of explaining phenomena is achieved when the I/O psychologist attempts to discover why
certain work behaviors occur.

Finding out that a company’s employee turnover rates are high because of employee dissatisfaction
with the levels of pay and benefits would be one example.

The goal of prediction would be addressed when a researcher attempts to use the scores from certain
psychological tests to predict which employee would be the best candidate for a management position,
or when a researcher uses a theory of motivation to predict how employees will respond to different
types of incentive programs

I/O psychology is also an applied science

Knowing this, an I/O psychologist might implement a program of increased employee participation in
company policy decision making in an effort to improve levels of employee job satisfaction.
Steps in the Research Process

The first step is the formulation of a problem or issue for study.

The second step is the generation of hypotheses.

The third step is designing the research, which includes selecting the appropriate research method or
design.

The fourth step is the actual collection of data, which is governed by the particular research design used.

The fifth step involves analyses of the collected data.

This leads to the final step, which involves the interpretation of results and the drawing of conclusions
based on the results.

Generation of Hypotheses

Researcher intends to measure, known as variables,

In the examples of research issues given earlier, job satisfaction, worker productivity, employee loyalty,
employment tenure, and absenteeism are all variables

hypotheses will later be tested through the analysis of the collected, systematic observations of
variables, better known as the collection and analysis of research data

Theory/Model the organization of beliefs into a representation of the factors that affect behavior

a researcher may eventually develop a theory or model, which is an organization of beliefs that enables
us to understand behavior more completely.

In social science, models are representations of the complexity of factors that affect behavior. In I/O
psychology, models are representations of the factors that affect work behavior.

developing a theory and diagramming that theory are convenient ways to organize our thinking and
our understanding of complex behavioral processes.

Collection of data - sampling

it is impossible to investigate all members of a particular population. For example

in pre-election polls of preferences, all potential voters cannot be surveyed. Instead, a sample is
selected, and the results obtained from this subgroup are generalized to the larger population.
Random sampling

For example, to choose a random sample of 20, workers from a company employing 200 workers, we
would begin with a list of all workers, and using a table of random numbers or a computer program that
generates random numbers, randomly select 20 workers.

Stratified sampling

For example, assume that 40% of the individuals in our total worker population are female and 60% are
male, and 25% are management and 75% are nonmanagement.

We would want to choose a sample that represented these percentages.

For example, imagine that a researcher decides to study the effects on work group productivity of two
managerial styles:

a directive style, whereby the manager closely supervises workers, telling them what they should be
doing and how they should be doing it, and a

nondirective, participative style, whereby the manager allows the workers a great deal of freedom in
deciding how they will get the work task done.

The researcher conducts the study on groups of directive and nondirective frontline managers who are
employed at 43 several factories that manufacture jet aircraft parts.

By collecting and analyzing data, the researcher concludes that directive managers lead more productive
groups.

However, the researcher might want to set some limits for the use of these findings.

The researcher might caution that these results may only apply to managers who are supervising factory
work groups and might not pertain to managers of service organizations, such as hospitals or
restaurants, to more creative jobs such as developing smartphone apps, or to managers of salespersons