________________________________________________________________________________

Subject PSYCHOLOGY

Paper No and Title Paper 16 : Community Psychology

Module No and Title Module 11 : Research Methods in Community Psyhchology -

Qualitative and Quanititative.
Module Tag PSY_P16_M11

TABLE OF CONTENTS
1. Learning Outcomes
2. Introduction
3. Qualitative Methods
3.1 Common Features of Qualitative Methods
3.1.1 Acts of Qualitative Research
3.2 Qualitative Research Methods
3.2.1 Participant Observation
3.2.2 Qualitative Interviewing
3.2.3 Focus Groups
3.2.4 Case Studies
4. Quantitative Methods
4.1 Common Features of Quantitative Methods
4.2 Quantitative Research Methods
4.2.1 Quantitative Description
4.2.2 Randomized Field Experiments
4.2.3 Nonequivalent Comparison Group Designs
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

4.2.4 Interrupted Time-Series Designs

5. Summary

1. Learning Outcomes
After studying this module, you shall be able to

 Identify the characteristics of Qualitative Research in Community Psychology

 Learn about strengths and limitations of some Qualitative Research Methods
 Identify the characteristics of Quantitative Research in Community Psychology
 Learn about strengths and limitations of some Quantitative Research Methods

2. Introduction
The study of psychology tends to be diverse, and draws out knowledge from other fields so as to
completely understand the psychological phenomena. Psychologists may carry out basic research
to further understand a particular area of interest in psychology, or may even conduct applied
research to solve problems in the clinic, workplace or other areas. In terms of Methodology,
Research in Community Psychology can be divided into two categories: Qualitative and
Quantitative. Though some suggest there is more affinity between Qualitative methods and
Community Psychology (Jarg B. Bergold, 2000) , we will discuss both these approaches.

3. Qualitative Research
“Research that investigates aspects of social life which are not amenable to quantitative
measurement. Associated with a variety of theoretical perspectives, qualitative research uses a
range of methods to focus on the meanings and interpretation of social phenomena and social
processes in the particular contexts in which they occur.” (Jupp, 2006 : Sage Dictionary of Social
Research Methods)The methods used in qualitative research, often in combination, are those
which are open-ended (to explore participants’ interpretations) and which allow the collection of
detailed information in a relatively close setting. These methods include depth interviewing,
ethnography and participant observation, case studies, life histories, discourse analysis and
conversational analysis. It is in the nature of qualitative research, with its emphasis on depth and
detail of understanding and interpretation, that it is often small-scale or micro-level. There is also
often a concern with the exploration of change and flux in social relationships in context and over
time.

3.1 Common Features of Qualitative Methods

Qualitative approaches are useful for examining situations, processes, and contexts that have not
been studied in detail. They give voice to perspectives that have not been fully articulated in
existing research. Thus, some community researchers use qualitative approaches in initial
exploration and theory development stages of a project, generating hypotheses that can be later
tested in quantitative research. But qualitative approaches also stand on their own, providing
detailed analysis of complex, dynamic, and meaningful lived experiences across a variety of
social and cultural contexts (Camic, Rhodes, & Yardley, 2003). This detailed analysis of
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

meaningful human experience in particular contexts is essential

for advancing scientific knowledge in community psychology (Tebes, 2005). Qualitative research
includes a diversity of methods, but most of them share the common features listed below.

1. Contextual meaning : The principal aim of qualitative research is to understand the meaning of a
phenomenon for persons who experience it in the contexts of their lives. This involves allowing
persons to “speak in their own voices” as much as possible, although interpretation by researchers
is also involved. Contextual understanding represents a form of “insider knowledge,” although it
is generated in part by discussions with outsiders (researchers).
2. Participant-researcher collaboration: Contextual meaning is created within personal, mutual
relationships that evolve over time between research participants and researchers. These methods
are thus especially apt for collaborative research with community members and for understanding
diverse social and cultural contexts. Participatory approaches extend modes of collaboration, so
participants contribute directly as members of the research team.

3. Purposeful sampling: The researcher develops a richly layered understanding of a particular

community group or setting. The sample of persons included in the research is usually small to
facilitate the level of detail needed. Researchers may also rely on their own experiences as
sources of information.
4. Listening: As much as possible, the researcher sets aside preconceptions and attempts to
understand the persons or setting on their terms, in their language and context. Attentiveness,
asking open-ended questions, and providing freedom for interviewees to structure their own
responses are preferred over standardized questionnaires (which often reflect researchers’
preconceptions or theories).
5. Reflexivity: Researchers also seek to be reflexive: stating their interests, values, preconceptions,
and personal statuses or roles as explicitly as possible—both to the persons studied and in the
research report. They also re-examine those assumptions in light of what they learn from the
research participants. This makes potential biases and assumptions as transparent as possible.
6. Thick description: Qualitative data in psychology are usually in the form of words. The
researcher seeks specific “thick description” of personal experiences, detailed enough to provide
convincing evidence of realism. This also affords later checking for significant details and
patterns. Other researchers can also use these detailed notes or transcripts to check the validity of
analysis and interpretation.
7. Data analysis and interpretation: The processes of data collection, data analysis, and
interpretation overlap, and the researcher moves back and forth among them. Analysis often
consists of identifying (coding) repeating themes or separating and comparing distinct categories
or stages. This framework promotes comparison of responses. Researchers can test the validity of
themes or categories by collecting and analyzing more data. Multiple coders and checks on inter-
coder agreement are used to strengthen reliability.
8. Checking: Usually after several rounds of refinement through data collection and analysis, the
researcher may check themes and interpretations by presenting them to informants or other
community members for correction, clarification, and interpretation. Participatory methods allow
community members to critically evaluate themes and challenge interpretations during the
analysis process.
9. Multiple interpretations: It is possible to have multiple interpretations or accounts of a topic.
However, an account should be internally consistent and compelling in terms of its realism and
thick description. Yet tensions and competing perspectives that arise from within a participatory
research team can also provide compelling evidence.
10. Generalization: Generalization of findings is less important than understanding meaning among

PSYCHOLOGY PAPER No. : 16 Community Psychology

MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

the persons sampled. Researchers may generalize findings by

identifying converging themes from multiple studies or cases.

3.1.1 Acts of Qualitative Research

How do qualitative researchers conduct a study? Catherine Stein and Eric Mankowski (2004)
identified four essential steps in qualitative study (they focus on qualitative interviewing). They
termed these acts that progress and build on one another over time, like acts in a play. (Figure 1)
As emphasized throughout this section, qualitative methods are very useful for attending to
unheard voices of marginalized groups and reducing (but not eliminating) power differentials
between researcher and participant. They also afford a deeper, contextual understanding of a
culture, community, or population. However, they do not simply express the views and voices of
participants. The assumptions and interpretations of the researcher are always involved, even with
open-ended questioning and discussion (Miller, 2004; Rapley & Pretty, 1999).

Figure 1 : Acts of Qualitative Research

3.2 Qualitative Research Methods

We next discuss four types of qualitative methods: participant observation, qualitative interviews,
focus groups, and case studies.

3.2.1 Participant Observation

Many community researchers, especially if they conduct participatory research, perform at least
some participant observation. It is a key component of ethnographic research in anthropology and
other social sciences. For some studies, participant observation is the primary method. Both
words in its title are important. Participant observation involves careful, detailed observation,
with written notes, interviews or conversations with citizens, and conceptual interpretation. It is
not just a description as the researcher becomes a member of a community or a collaborator in its
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

efforts, an actor in community life. This provides at least some

of the experiential insider knowledge of community members, while the researcher also strives to
maintain something of the outsider perspective.

Strengths and Limitations : Participant observation is the method of choice for a researcher
seeking maximum insider knowledge and depth of experience in a community. The participant
observer knows the setting thoroughly and can communicate its essence vividly. This method also
maximizes the researcher community relationship and affords thick description of many aspects
of community life.

However, that depth of knowledge comes at a price. First, the focus on one setting necessarily
means that generalizability to other settings is a problem. This can be reduced by visiting other
settings—usually in less depth but long enough to discern the applicability of one’s findings. A
second issue concerns whether the researcher’s experiences and records are representative of the
setting and its dynamics. The researcher’s notes, analysis, and interpretation can be affected by
selective observation, selective memory, and selective interpretations. Findings can also be
affected by an unrepresentative sample of informants or of situations studied (e.g., observations
of formal meetings but not informal caucuses or personal contacts). Researchers need to explicitly
report their value commitments relevant to the study, and whether they took sides in a
controversy, so readers can judge the effects of these choices on data collection and
interpretation. Another problem is that the researcher is influencing the phenomena or community
under study. Field notes and interpretations should explicitly indicate the extent of the
researcher’s influence on the actions of others so the impact of the researcher’s participation can
be assessed. A final limitation is the role conflict created by playing both participant and observer
roles (Wicker & Sommer, 1993). An ethical and personal problem concerns what the researcher
tells the members of the community about the research. Playing a role that is both an insider and a
researcher can be stressful. Striking a balance between these is an important part of gaining entry,
forming a relationship, and seeking to benefit the community.

3.2.2 Qualitative Interviewing

Interviewing a sample of individuals has become a popular qualitative research format in

community psychology. The interview is often open-ended or minimally structured to promote
participants’ describing experiences in their words. Samples are small to facilitate interviewing
and analysis in depth. The researcher is not necessarily a participant in the community under
study but usually does assume a role of collaboration or extended contact with interviewees.

Strengths and Limitations : Qualitative interviewing allows flexible exploration of the

phenomenon of interest and discovery of aspects not anticipated by the researcher. It is based in a
strong relationship between researcher and participant. It involves attending to the voices of
participants and thick description of their experiences. It can challenge the researcher’s
preconceptions and affords contextual understanding of a community, culture, or population.

Interviewing has several advantages over participant observation. Data collection can be more
standardized, limiting biases of selective perception, memory, and interpretation. Interviews can
be recorded and transcripts prepared so analysis can be based on participants’ actual words.
Analysis can also be standardized and performed by multiple, independent raters, not just the
interviewer, which increases reliability and validity. The interviewer can develop a relationship
with the setting and participants that is mutual and trusting yet with less role conflict than
participant observation. Of course, all these also mean that the insights developed from interviews
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

are less direct than those from participant observation. These

advantages require intensive study of a small sample, which means that generalizability of
findings is often limited. Also, the time required for research interviews may subtly exclude
participants in marginalized groups or demanding circumstances. Differences of interpretation
between participants and researchers do create challenges.

3.2.3 Focus Groups

A focus group discussion is an interview with a group. It generates thick description and
qualitative information in response to questions or discussion topics posed by a moderator. Using
focus groups, researchers can assess similarities and differences among individuals and allow
participants to elaborate on ideas and themes by reacting to each other, not just to an interviewer.
In focus group research, the group, not the individual, is the unit of analysis: The sample size is
one for each group. Individual comments are not independent of other group members; indeed,
one of the purposes of the focus group is to elicit discussion. Each group is usually composed of
6–12 participants who share some characteristic of concern to the researchers—for example, the
same race, gender, culture, or age, similar occupations, or the same health problem. This
homogeneity helps to promote free discussion and ability of participants to be able to identify
with each others’ experiences. A group of strangers is preferred to minimize the effects of prior
personal contacts. Multiple focus groups are needed to provide broader information and to
compare populations. However, as with qualitative interviewing, samples are seldom
representative of a large population. The goal is to generate contextual understanding. The
moderator’s responsibilities include creating an environment conducive to free discussion,
speaking in language comfortable to all participants, ensuring that all members participate,
eliciting both agreement and disagreement, and balancing between being nondirective and
covering all topics of interest to the researchers. The moderator uses a discussion guide that
includes topics to be discussed and that moves from general topics to specific phenomena
relevant to the research. Analysis of focus group data is similar to the process of analyzing
individual qualitative interviews.

Strengths and Limitations : These are similar to qualitative interviewing. However, focus
groups have several advantages over other qualitative methods. Researchers can structure
discussion and learn about topics of interest and personal experiences of others more easily than
with participant observation. Compared to individual interviews, focus groups allow greater
access to shared knowledge and mutual discussion. They also allow researchers to observe social
interaction among group participants, perhaps revealing behavioral patterns unavailable in
individual interviews. However, a focus group moderator has less flexibility to ask for
elaboration, control changes of topic, or learn about individuals in depth than an interviewer of
individuals. Focus groups are not a good approach for understanding an individual’s unique
experiences and cannot simply be used as a substitute for individual interviewing. Focus groups
are especially useful for gaining cultural understanding. They also can help explore a topic or test
questionnaires prior to quantitative studies.

3.2.4 Case Studies

The case study method, usually conducted on individuals in clinical psychology, can be applied to
an organization, locality, or change process (Bond & Keys, 1993; Mulvey, 2002; Evans, Hanlin,
& Prilleltensky, 2007). Community psychologists also can study an individual in relation to the
settings in that person’s life (Langhout, 2003). They may conduct multiple case studies so
comparisons can be made. For example, Wasco, Campbell, and Clark (2002) interviewed eight
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

advocates who worked with rape victims, about how the

advocates coped with the emotional trauma encountered in their work, and about personal and
organizational resources that promoted coping.

Case studies provide a bridge connecting qualitative and quantitative approaches. A case study
may rely on any or all of the qualitative methods we have described. It may also use qualitative
archival data (i.e., from archives or records) such as minutes of group meetings, organizational
policy manuals, or newspaper stories. Archival data can also be quantitative records such as
police statistics, records of attendance at programs, or quantitative evaluations of whether a
program attained its goals. Case studies also can use other quantitative measures, such as
questionnaires.

Strengths and Limitations Like participant observation, a case study can examine in depth a
single person, setting, or locality. Case studies are excellent for understanding the nuances of
cultural, social, or community contexts. They can afford thick description and contextual
understanding. By using multiple data sources, subjective biases can be checked. The longitudinal
focus of most case studies is also useful. Although case-study researchers cannot study causes
and effects with experimental control, they can identify complex patterns of causation in natural
settings. Their focus on a single case is also the principal limitation for case study methods.
Generalizability of findings to other settings is uncertain. Researchers can include multiple case
studies in one analysis, but that may weaken some of the strengths described here. Sometimes,
description and analysis of one case is the goal, not generalization to other cases. Involvement in
the setting or locality studied may create insider-outsider role conflicts. The use of archival
records presents both advantages and problems. Written records can provide information on
meetings or other events not attended by the researcher and remembered imperfectly by interview
informants. Archival records can also document events in the history of an organization or
community. However, researchers who review archival data may not discover the processes that
they are most interested in. For example, conflict and compromise preceding a group decision are
usually omitted or sparingly recorded in meeting minutes.

4. Quantitative Methods
“Research involving the collection of data in numerical form for quantitative analysis. The
numerical data can be durations, scores, counts of incidents, ratings, or scales. Quantitative data
can be collected in either controlled or naturalistic environments, in laboratories or field studies,
from special populations or from samples of the general population. The defining factor is that
numbers result from the process, whether the initial data collection produced numerical values, or
whether non-numerical values were subsequently converted to numbers as part of the analysis
process, as in content analysis”(Jupp, 2006). Changes over time can be more easily tracked using
quantitative methods, as measures of the same properties can be taken at several points during an
intervention. For example, if a community project is intended to engage young people, comparing
those inclined towards antisocial behaviour with those who are not, measures of shoplifting,
juvenile arrests, nuisance calls to the civil services, such as fire and ambulance, can be used
across the time of the intervention in order to establish the efficacy of the project. It would be
very difficult to establish actual effects without numerical data. Quantitative methods are
particularly useful in helping us describe and model the multileveled influence of environmental
factors on individual health and well-being (Luke, 2005).

4.1 Common Features of Quantitative Methods

PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

A great diversity of quantitative methods exists. However, most

quantitative methods in community research share some common features. We do not wish to
repeat all of what you may have learned in previous methodology courses, so the list below
focuses on features that offer clear contrasts with qualitative methods and highlights how
quantitative methods can be adapted to community research.

1. Measurement and comparisons: The principal aim of quantitative methods is to analyze

measurable differences between and the strength of relationships among those variables. They
facilitate understanding variables, predicting outcomes, and understanding causes and effects.
Quantitative research generates “outsider knowledge” that affords comparisons across contexts.
2. Numbers are data: Although some variables are categorical (e.g., an experimental program
compared to a control group), the purpose is almost always to study their relationship to measured
variables. While researchers using qualitative methods look for patterns in words and narratives,
researchers using quantitative methods seek patterns in numbers.
3. Cause and effect : One important objective is to understand cause-effect relationships. This can
then lead to the prediction of consequences and inform social action to promote desirable
changes. Experiments and similar methods are often used to evaluate the effects of social
innovations, programs, or policies. Even non-experimental quantitative studies identify empirical
relationships that can eventually lead to knowledge of causes and effects and to social innovation.
4. Generalization : Another important objective is to derive conclusions that can be generalized at
least to some extent across contexts, settings, and communities (e.g., empirical findings showing
that a prevention program or social policy is effective in many communities).
5. Standardized measures : Standardized measurement instruments are preferred to ensure
reliable, valid measurement. The flexibility and contextual sensitivity of qualitative methods are
lost, but the comparability of findings across studies and control of extraneous variables are
increased. Next, we discuss four specific types of quantitative methods in community
psychology research: quantitative description, randomized field experiments, nonequivalent
comparison group designs, and interrupted time-series designs.

4.2 Quantitative Research Methods

These are only some of the available quantitative methods for community research.

4.2.1 Quantitative Description

Quantitative description methods include a variety of procedures—e.g., surveys, structured

interviews, behavioral observations of community settings, epidemiological studies, and use of
social indicators (i.e., census data, crime and health statistics). They are quantitative but not
experimental, and they do not involve manipulation of an independent variable. They can be used
for such purposes as the following:

 To compare existing groups (e.g., women’s and men’s perceptions of crime)

 To study associations between survey variables (e.g., correlation of family income with health or
changes over time in adolescent sexual attitudes)
 To measure characteristics of community settings (e.g., measure the frequency of emotional
support and giving advice in mutual help groups)
 To conduct epidemiological studies to identify factors predicting the presence or absence of an
illness (e.g., behaviors that increase or decrease the risk of HIV infection)
 To study relationships between geographic-spatial and social environments (e.g., the correlation
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

between density of liquor stores and crime rates in

neighborhoods)

Statistical analyses may include correlation, multiple regression, path analysis and structural
modeling, and even t tests and analyses of variance to compare naturally occurring groups. These
studies may be cross-sectional, sampling only one point in time, or longitudinal, sampling
repeatedly over time. Quantitative description usually samples more individuals than either
qualitative studies or experiments. This facilitates statistical analysis and generalizability. To
enable a study of this breadth, these methods rely on previous knowledge and/or exploratory
research to determine which variables to study, how to measure them, and whom to sample.

Correlation and Causation : Early undergraduate education in psychology typically contrasts

correlation and causation. Just because two factors are associated statistically does not mean that
one causes another. The causation could just as easily run in the opposite direction than what you
think (B causes A rather than A causing B). Or the causal factor may be a “third variable” that
determines both correlated variables (C causes both A and B). However, under some conditions,
non-experimental designs can be used to identify causal patterns and test causal hypotheses. The
simplest case involves precedence in time: If the change in A is correlated with the change in B
but A consistently precedes B, a causal interpretation (A causes B) is more warranted (although a
third variable still may be involved). A theoretical model, based on prior knowledge of
relationships among A, B, C, and other related variables, strengthens causal inference from non-
experimental data. Such causal inference relies on logic models and/or statistical control of
extraneous variables, not experimental control.

Community Surveys : Surveys of community samples, using standardized questionnaires or

other measurements, are quantitative description methods. Community surveys can focus on
organizations as the unit of analysis. Community coalitions bring together representatives of
various segments of a locality to address an issue such as domestic violence or promoting positive
youth development .

Epidemiology : These methods are useful for community research concerned with health and
mental health. Epidemiology is the study of the frequency and distribution of disorders, and of
risk and protective factors for these. Epidemiology is usually a precursor to more experimental
studies of causal factors of these disorders and is essential to practical planning of prevention and
treatment. Epidemiology is most often used in the discipline of public health but is also used in
the social sciences (e.g., Mason, Chapman, & Scott, 1999).

Two basic epidemiological concepts are incidence and prevalence. Incidence is the rate of new
occurrences of a disorder in a population within a specific time period (usually a year). It is thus a
measure of the frequency of the onset of a disorder. Prevalence is the rate of existing occurrences
of a disorder in a population within a time period. It includes new cases and continuing cases of
the disorder that began before the time period studied. Both concepts are usually expressed as
rates (e.g., the number of cases per thousand persons in the population). The incidence–
prevalence distinction is important for community psychology. Prevention is more concerned
with incidence—the frequency of new cases. Prevalence—the rate of existing cases—is relevant
to mutual help or mental health services policy. When incidence and prevalence rates have been
determined for a population, epidemiological research is focused on identifying risk and
protective factors. Risk factors are associated with increased likelihood of a disorder. These may
be causes of the disorder or simply correlated with it. Exposure to stressors or lack of coping
resources are examples of risk factors. Protective factors are associated with lesser likelihood of
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

a disorder; they may counteract or buffer the effects of the

disorder’s causes or simply be correlated with other factors that do so. Personal or cultural
strengths and support systems are protective factors.

Mapping Physical and Social Environments

The increasing availability of geographical information systems (GIS) methods offer a rich new
resource for studying relationships between physical-spatial aspects of communities and their
psychosocial qualities (Chirowodza, van Rooyen, Joseph, Sikotoyi, Richter, & Coates, 2009;
Luke, 2005). GIS methods can be used to plot onto a map any data available for spatial locations.
Archival data sources can include census information on population density or average household
income, or social indicators such as neighborhood crime rates or density of liquor stores.
Community survey data can also be entered in GIS databases if associated with respondents’
residences (Van Egeren, Huber, & Cantillon, 2003). GIS data and the resulting maps can be used
for quantitative statistical analysis or for visual searching for spatial patterns (a more qualitative
approach).

Strengths and Limitations : Quantitative description methods have a number of strengths.

Standardized measurement affords statistical analysis and large samples that provide greater
generalizability. These methods can be used to study variables that cannot be manipulated in an
experiment. Epidemiological research can be used to identify risk and protective factors and
evaluate the outcome of preventive efforts. Finally, these studies often identify factors that can be
targeted for social or community change, even without experimental knowledge of specific
causes and effects. For example, one need not know all the cause-effect relationships for youth
violence in order to identify risk and protective factors and to initiate change efforts.

However, these methods have several limitations. They rely on prior knowledge to select and
measure variables and populations. Also, except for GIS approaches, the knowledge provided by
these studies is usually “decontextualized”—gathered from individuals but not associated with
existing settings, communities, or cultures. This can increase breadth of sampling but limits
knowledge of these contextual factors. The study of causes and effects is limited, as we have
discussed. The focus of epidemiological research on disorders also limits its utility for community
psychology (Linney & Reppucci, 1982). Community psychology is concerned with overall
psychological well-being, including but not limited to disorders. When mental disorders are
studied, difficulties of accurate diagnosis and measurement make its epidemiology more difficult
than that of physical disease. Also, community psychology’s focus on promotion of strengths
includes identifying protective factors for disorders but goes beyond that to concern the
development of positive qualities.

Now we will discuss three research methods as forms of experimental social innovation.

4.2.2 Randomized Field Experiments

This is the most rigorous form of experimental social innovation. Participants (individuals or
settings) are randomly assigned to experimental or control groups. These are compared at a
pretest before the implementation of the experimental social innovation, at which time they are
expected to be equal on measures of dependent variables. They are compared again at post-test(s),
when they are expected to differ because of the effects of the innovation. Follow-up post-tests can
continue over several years. The experimental social innovation represents the experimental
condition. The control condition can often be “treatment as usual” under existing policy or
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

practices. Another experimental approach is to compare two

different innovations with each other, such as two contrasting prevention programs in a school
(Linney, 1989). A third approach is to provide the experimental innovation to members of the
control group after the posttest. They serve first as a control group and then receive the
innovation, minimizing ethical problems with their not receiving it originally.

A key issue is the method of assignment to experimental and control conditions. If this is random,
many confounding variables are controlled. These include individual differences in personality,
coping skills, social support networks, and life experiences that may affect their responses to the
innovation. Confounds also include differences between groups in demographics such as gender,
age, race, culture, and family income. In the laboratory, random assignment is taken for granted,
but in the community, it must be achieved by collaboration and negotiation with community
members (Sullivan, 2003).

Strengths and Limitations Randomized field experiments are unsurpassed for clarity of cause-
effect interpretation (e.g., for testing effects of a social innovation). With greater control over
confounding factors, researchers can make more confident interpretations of its effects.
Moreover, if experimental studies demonstrate the effectiveness of a social innovation, advocacy
for it can be more effective. For example, randomized field experiments have helped to document
the effects of many preventive interventions and increase the credibility of prevention efforts
generally (Weissberg, Kumpfer, & Seligman, 2003).

However, experiments require substantial prior knowledge of the context to propose social
innovations worth testing and to choose measurements. A useful sequence might be to conduct
qualitative studies to understand the context and key variables, then quantitative descriptions to
specify risk and protective factors and refine measurements, then developing an experimental
social innovation and conducting an experiment to evaluate its effectiveness. Even during the
experiment, qualitative or individualized quantitative methods are helpful to understand
differences in outcomes and whether these were due to the intervention (Lipsey & Cordray,
2002). The intrusiveness of experiments also raises issues of control in community settings.
Permission is needed to collect quantitative data (often in multiple waves) and to randomly assign
participants to experimental conditions. Those decisions must be explained and negotiated with
community members.

4.2.3 Nonequivalent Comparison Group Designs

For a variety of reasons, many settings simply cannot support random assignment to experimental
and control conditions. For example, seldom can a school randomly assign some children to an
innovative classroom and others to a control classroom. Even if they did, the students may mix at
lunch or recess so much that the independence of experimental and control conditions is greatly
reduced. Providing the innovation to all students in a grade and comparing their outcomes to
another school or to students in a previous year means, of course, that assurance of equivalence
between groups is lost. Comparing a sample of schools rather than a sample of individuals
(making the unit of analysis the school, not the individual) may be prohibitively expensive. Yet
many of the strengths of the experiment can be retained if researchers are creative about working
around such obstacles. Using a nonequivalent comparison group is a common approach.

Nonequivalent comparison group designs are used whenever assignment to experimental or

comparison condition is something other than random. For example, different classrooms within a
school or different schools within a region may serve as experimental and comparison conditions.
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

Assignment to classroom or choice of school is not random, but

the classes or schools may be similar. The choice of comparison group is critical to generating
interpretable results. In schools, student socioeconomic status, race, gender, and age are examples
of variables to equate as much as possible in the two groups. Teacher demographics, school size,
and curriculum also need to be similar (Linney & Repucci, 1982).

Strengths and Limitations : Using an existing group as a comparison condition is practical and
less intrusive than randomized experiments. However, the control of confounding factors is much
weaker, and clarity of interpretation and confidence in conclusions is decreased. Researchers in
this situation must collect as much data as possible on factors that may confound the comparison.
This allows them to document the similarity of the two conditions or to use those variables as
statistical controls. For example, researchers may be able to show that the average family income
of the experimental and control conditions was similar or control effects of family income
statistically. The ultimate goal is to weaken or eliminate plausible competing explanations for
findings.

4.2.4 Interrupted Time-Series Designs

Another approach is the use of interrupted time-series designs. In the simplest case, this involves
repeated measurement over time (a time series) of a single case (an individual, organization,
locality, or another social unit). In an initial baseline period, the participant or setting is monitored
as measurements of dependent variables are collected. This provides the equivalent of a control
condition. Then, the social innovation (e.g., program, policy) is introduced while measurement
continues. Data collected in the baseline period are compared to data collected during or after the
innovation was implemented. This is termed an interrupted time-series design because the
innovation interrupts the series of measurements. This approach combines time-series
measurement with an experimental manipulation, providing a useful design for small-scale
experimental social innovation when a control group is not available.

Strengths and Limitations : Time-series designs are practical. They also afford understanding of
change over time in a specific context, such as one community, while standardizing measurement
and minimizing extraneous confounds. However, a time-series design with one group is still open
to a number of external confounds (Linney & Reppucci, 1982). These include seasonal or cyclical
fluctuations in the variables measured. An example is that the number of college students who
seek counseling rises as final exams approach. If seeking counseling is used as a dependent
variable in a time-series study, researchers must take this seasonal rise into account. A further
confound concerns historical events that affect the variables measured. Finally, findings from a
single case or community, even over a long time period, may not generalize to other
communities. A key issue for time-series designs is the number and timing of measurements in
the baseline and experimental periods. Social innovations may have gradual or delayed effects
difficult to detect in a short time-series design. Seasonal or cyclical fluctuations (confounds) in
the dependent variable may be detected if the time series is long enough.

Multiple-Baseline Designs This is a form of interrupted times-series design that reduces the
problems of external confounds and generalizability. Think of this design as a set of time-series
studies, each conducted in a different community and compared to each other. The experimental
social innovation is implemented at a different time in each community so effects of an external
historical factor (happening at the same time for all communities, such as national publicity about
tobacco use) will not be confounded with the innovation. If measures of the dependent variable
show a change soon after the implementation of the innovation, at a different date in each
PSYCHOLOGY PAPER No. : 16 Community Psychology
MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

community, confidence can be stronger that the innovation

caused this effect. This also provides some evidence of generalizability. In effect, the design tests
whether findings from one community can be replicated in other communities within a single
study. The multiple-baseline design combines the strengths of the interrupted time-series and
nonequivalent comparison group designs. However, the multiple communities studied are still
nonequivalent (assignment of individuals to them is not random), and differences among them
still exist that complicate interpretation. Still it is a very useful way to combine repeated
measurement, contextual study of a single community, and replication across communities.

5. Summary
These two summary tables, (A and B) encapsulate the key themes in this chapter. Table A :

Qualitative Research Quantitative Research

Methods include focus groups, in-depth Surveys, structured interviews & observations, and
interviews, and reviews of documents for types reviews of records or documents for numeric
of themes information
Primarily inductive process used to formulate Primarily deductive process used to test pre-specified
theory or hypotheses concepts, constructs, and hypotheses that make up a
theory
More subjective: describes a problem or More objective: provides observed effects (interpreted
condition from the point of view of those by researchers) of a program on a problem or condition
experiencing it
Text-based Number-based
More in-depth information on a few cases Less in-depth but more breadth of information across a
large number of cases
Unstructured or semi-structured response options Fixed response options
No statistical tests Statistical tests are used for analysis
Can be valid and reliable: largely depends on Can be valid and reliable: largely depends on the
skill and rigor of the researcher measurement device or instrument used
Time expenditure lighter on the planning end and Time expenditure heavier on the planning phase and
heavier during the analysis phase lighter on the analysis phase
Less generalizable More generalizable

PSYCHOLOGY PAPER No. : 16 Community Psychology

MODULE No.: 11 Research Methods in Community Psychology
________________________________________________________________________________

Table B : Comparison of Qualitative Research Methods

PSYCHOLOGY PAPER No. : 16 Community Psychology

MODULE No.: 11 Research Methods in Community Psychology