Design, Methods and Measurement for Applied Research

1
Design, Methods and Measurementfor Applied
Research
John J. Green, Ph.D. Institute for
Community-Based Research Division of Social
Sciences/Center for Community and Economic
Development Delta State University Asset
Mapping, Needs Assessment and Evaluation COD
695 Summer II, 2005
2
Basic Research Designs
Case Design A study that focuses on one case (a
case being a unit for research which could
include a person, group, organization, community,
etc.) within its social context. Cross-Sectional
Design A study in which data are collected for
all of the variables of interest using one sample
at one time. Longitudinal Design A study where
data are collected at least two times, includes
what are known as trend, cohort and panel
studies. Experimental Design A study in which
the independent variable(s) are controlled in
some way. Particularly useful in explanatory and
causal focused research. Often involves the use
of a control group. Quasi-Experimental Design A
study missing one or more aspects of a true
experimental study, most frequently random
assignment into experimental and control groups.
3
Qualitative Research
Qualitative research is conducted to develop an
in-depth, descriptive and interpretive
understanding of the social world. Qualitative
research methods are used in studies that are
exploratory in nature and those where the
researcher wants to develop a detailed and
in-depth account of a particular
issue. Qualitative methods may be used for
theory development and theory elaboration.
Three often used qualitative research methods
observation, key-informant interviews and focus
groups.
4
Observation
The purpose of observation is to collect
information about the social world in what is
considered its (relatively) more natural state.
The researcher is attempting to understand things
from the perspective of how they actually take
place. There are two primary forms of
observation Non-participant/bystander
observation The researcher is present as a
witness, but s/he is not engaging in group
activities. Participant observation The
researcher does engage in group activities as a
way to experience the world.
5
Key-Informant Interviews

• Key-informant interviews are generally
  qualitative, in-depth discussions guided by a set
  of open-ended questions.
• They may also include some closed-ended
  questions.
• They are conducted with diverse persons who have
  knowledge and interest in the topic(s) of
  interest.

6
Focus Groups
Group process research involves the use of
groups of people interacting to generate
data. Focus groups represent one form of group
process research that uses group interaction and
discussion on a set of topics as the primary mode
of data construction. It is advisable to conduct
multiple focus groups for any particular research
effort. This allows for greater participation
(and thus wider representation of results) and
the opportunity to compare results across
groups. Similar to key-informant interviews,
focus groups require a list of well-defined
topics and questions to guide discussion.
7
It is usually not enough to have participants
just answer questions.
To stimulate thought and participation, ask them
to write things down and report back, draw
diagrams, role play, complete a short survey, etc.
8
Quantitative Research
In quantitative research, our attention is
shifted to the level and/or amounts of social
phenomena. Quantitative research often focuses
attention on the patterns among and between
phenomena as expressed through statistical
summaries. May be used for descriptive purposes
and theory testing. Here, we will look at two
common forms of quantitative research analysis
of secondary data and surveys.
9
Secondary Data
Data collected by another party for a different
or broader purpose may be useful for
community-based research. Examples include
the US Census Census of Agriculture CDC and
Health Department Reports American Housing
Survey General Social Survey
10
Surveys
A method of research where data are obtained from
responses to a standardized questionnaire. Strate
gies for conducting surveys include Mail Drop-o
ff/pick-up Group administered In-person/face-to-
face Telephone (example Delta Rural
Poll) Web-based
11
Measurement
Measurement The classifying of units of analysis
by categories to represent a variable
concept. Four basic levels of measurement/types
of variables
Ratio Similar to interval, but with the added
feature of a common sense zero
Interval Adjacent ranked categories are of equal
distance/value
Ordinal Mutually exclusive categories, order is
ranked in a logical manner
Complexity
Analysis Options
Nominal Descriptive, mutually exclusive
categories, order is incidental
12
Composite Measures
Scale Measurement tool used to assess a single
dimension of a construct composed of multiple
indicators. Example Household food security
scale (adapted from Bickel, Nord, Price, Hamilton
and Cook 2000). .
I am going to read you statements that people
have made about their food situation. Please tell
me whether the statement was OFTEN, SOMETIMES, or
NEVER true for you or other members of your
household in the last 12 months.
13
Composite Measures (contd)
Index Measurement tool used to assess the extent
of a construct (may include multiple
dimensions). Example United Nations Human
Development Index (UN 2005).
The human development index (HDI) is a composite
index that measures the average achievements in a
country in three basic dimensions of human
development A long and healthy life, as
measured by life expectancy at birth Knowledge,
as measured by the adult literacy rate and the
combined gross enrolment ratio for primary,
secondary and tertiary schools A decent standard
of living, as measured by GDP per capita in
purchasing power parity (PPP) US dollars. The
index is constructed using indicators that are
currently available globally.