U

Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom

mMM SS

CHAPTER 2

RESEARCH IN COMMUNICATION – A BRIEF THEORETIC OVERVIEW

2.1 INTRODUCTION

The purpose of this chapter is to provide a brief theoretic overview of some

aspects relevant to research in communication. The researcher believes that
government communicators always need to conduct research on the basis of
sound theoretical guidelines and principles to ensure research of good quality.
Writing about Total Quality Management (TQM), Pace (2001) defines quality as
“doing the right thing the right way the first time and every time”, and further
defines “right” and “wrong” with regard to quality as follows:

The right thing must be understood from both internal (product/service) and
external (customer) perspectives. This means that the product or service meets
customer requirements, performs as stated, is priced fairly, and is delivered on
time.

The right way is the most effective, most efficient, lowest cost, fastest, highest
value approach to producing the right outcome the first time and every time. It
applies conformance to all applicable standards and specifications as well as
minimisation of the costs of poor quality such as rework, waste, and scrap.

Pace (2001) concludes that “poor quality could be defined as either doing the
wrong thing (or failing to do the right thing) or doing the right thing the wrong way
(or failing to do the right thing the right way every time).”

12
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

The researcher argues that communication research is the “right thing” to do, but
that it is of no use if not done in “the right way.”

In chapter 2 communication research is defined. The researcher furthermore

distinguishes different types of research, discusses the different steps in the
research process typical to most research projects and refers to some of the
challenges of communication research. Lastly, a few points regarding the use of
communication research by governments are introduced.

2.2 COMMUNICATION RESEARCH DEFINED

Definitions and descriptions of different focus areas in research and of various

research methodologies and processes are more readily available than a
definition of ‘research’, and Leedy (1997:3) remarks that “the word research is
used in everyday speech to cover a broad spectrum of meaning, which makes it
a decidedly confusing term …”

Reinard (2001:3) provides the following short definition of research: “Research is

the systematic effort to secure answers to questions.” He expands on this
concise definition by stressing the point that “these questions are not mundane
ones”, but that “research questions deal with issues requiring reference to data
and information” (Reinard, 2001:3-4). Leedy (1997:3) defines research as “the
systematic process of collecting and analysing information (data) in order to
increase our understanding of the phenomenon with which we are concerned or
interested.” Powell (1997:2) does not provide his own definition of ‘research’, but
quotes three definitions from others:

“Studious inquiring or examination; especially: investigation or experimentation

aimed at the discovery and interpretation of facts, revision of accepted theories

13
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

or laws in the light of new facts, or practical applications of such new or revised
theories or laws” (Webster’s Seventh New Collegiate Dictionary, cited by Powell,
1997:2).

“A method or study by which, through the careful and exhaustive investigation of

all the ascertainable evidence bearing upon a definable problem, we reach a
solution of that problem” (Hillway, cited by Powell, 1997:2).

“Research is best conceived as the process of arriving at dependable solutions to

problems through the planned and systematic collection, analysis, and
interpretation of data” (Mouly, cited by Powell, 1997:2).

Both interesting and useful to understand the meaning of research, is Leedy’s

guidelines as to what research is not (Leedy, 1997:4):
• Research is not mere information gathering.
• Research is not mere transportation of facts from one location to another.
• Research is not merely rummaging for information.
• Research is not a catchword to get attention.

According to Powell (1997:2) “there is no one definition of research, in part

because there is more than one kind of research.”

Focus areas in research covered extensively in the literature available, include

the following:
• marketing research (Martins, 1996a:3-22; Chisnall, 1991:6; Crimp,
1990:3; Bailey, 1982:2)
• market research (Cooper, 1998:1015-1024); Fairweather, 2001; Smith,
1998:29-65; Yule, 2001; Shalofsky, 1998:1103-1128)

14
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

• social research (Bless and Higson-Smith, 2000; Bulmer, 2000a:8-9;

Groenewald, 1981; Möhler, 1998:1025-1032; Shalofsky, 1998:1103-
1128)
• advertising research (Hansen, 1998:653-724; Yasuda and Spence,
2000:179-201; Martins, 1996b:550-567)
• public opinion research (Khoury, 1989; Mattes, 1993:30-32; Worcester,
1999; Taylor, 1998:975; ESOMAR, 2000a)
• media research (Raimondi, 1998:803-838; Martins, 1996c:570-597).

Similar to both Reinard and Leedy, Chisnall (1991:6) and Weiers (1984:2) refer
to research as being a “systematic process” when they define marketing
research. Chisnall (1991:6) defines it as being “concerned with the systematic
and objective collection, analysis and evaluation of information about specific
aspects of marketing problems in order to help management make effective
decisions.” Weiers (1984:2) adapts a definition presented by Kotler in 1980 in
the following way: “Marketing research is the systematic design, collection,
analysis and reporting of data and findings relevant to a specific marketing
situation.”

In this chapter, the researcher quotes from different sources consulted where the
focus was either on one or more of the focus-areas mentioned above – e.g.
marketing research and social research. The researcher is of the opinion that
the same broader principles, processes and challenges apply to communication
research and the other focus areas identified.

Analysis of definitions available, leads the researcher to the conclusion that most
authors emphasise the ‘why’ and ‘how’ when defining research and the different
focus areas in research.

15
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Referring specifically to the ‘how’, Bless and Higson-Smith (2000:11) defines

scientific research as “the translation into practice of the relationship between
facts and theory … in order to acquire specific information” and distinguish the
following four characteristics of scientific research:
• “Scientific research is empirical since the aim is to know reality. Each step
is based on observation, be it when collecting the basic facts or when
testing the explanation, assessing the value of the prediction or the result
of an intervention.
• Scientific research is systematic and logic. Not only must the observation
be done systematically but a certain logical order must be followed all
along (see discussion regarding research process, paragraph 2.4).
• Scientific research is replicable and transmittable. Since the observation
is objective and the explanation logical, anyone placed in exactly the same
circumstances can observe the same event and make the same
reasoning, leading to the same explanation and prediction. Moreover, it is
possible to communicate each step of the research and to transmit the
acquired knowledge.
• Scientific research is reductive. To grasp the main relationships of laws,
the complexity of reality is reduced. All details which are not essential or
which have little influence on the process under investigation are omitted”
(Bless and Higson-Smith, 2000:5-6).

The only definition of communication research available to the researcher is that

of Reinard (2001:5) who defines communication as “the process by which
participants transact and assign meaning to messages” and explains that a
message is “the set of verbal and non-verbal cues communicators exchange.”
According to Reinard (2001:4), communication research is “a speciality that
studies message-related behaviour.” He explains that “some people have
difficulty separating communication research from work in psychology, sociology

16
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

or literature”, and that they argue “that since ‘meanings are in people’ (Berlo,
cited by Reinard, 2001:4) any study of people is communication research.”

Relevant for the communication researcher is the reminder by Worcester

(1999:3) who, writing about public opinion research, says that “polls do not
measure some abstract ‘truth’, but people’s perceptions.” Worcester (1999:3)
refers to the wisdom expressed by Epictetus as long ago as the first century that
“perceptions are truth, because people believe them.”

Although researchers are confronted with various challenges when conducting

communication research (see paragraph 2.5), and the reality that contemporary
communication testing is still far from perfect, Hansen (1998:716) argues that it is
still good enough to warrant the recommendation: “Test rather than guess.”

The researcher fully agrees with Hansen’s argument, but wants to emphasise the
necessity of conducting this “test” in the “the right way.” Furthermore, the
researcher is of the opinion that the spectrum of aspects covered by
communication research conducted by governments is broader than “studies in
message related behaviour” referred to by Reinard in his definition of
communication research as quoted. Research conducted by governments to
enhance the effectiveness of government communication and the dissemination
of government information (see chapers 3 to 5 of this dissertation) also includes
studies related to communication and information products and initiatives of
governments; exposure to and attitude towards the different mediums that can
be used for government communication and the dissemination of government
information; awareness and knowledge of and attitude towards government
performance and towards government policies and initiatives on a wide spectrum
of issues. To some extent, therefore, the research conducted by governments to
enhance the effectiveness of government communication and the dissemination
of government information focuses on more than merely communication and

17
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

information. The focus includes areas like media research, public opinion
research and social research.

2.3 TYPES OF RESEARCH

2.3.1 Introduction

There are many different ways according to which research studies are classified
– e.g. according to various focus areas (see paragraph 2.2), the environment
from which the research is conducted (e.g. academic, business or government)
or the specific technique of data collection (e.g. personal interviews, telephone
interviews or mailed questionnaires).

The researcher distinguishes different types of research according to the

classification of Bless and Higson-Smith (2000:37-44). These authors suggest
that research studies can be classified according to the following three aspects:
• the methodology used (quantitative and qualitative research)
• the reasons for the research being conducted (basic social research and
applied social research)
• the demands of the research question (exploratory research, descriptive
research, correlational research and explanatory research).

2.3.2 Different types of research

2.3.2.1 Quantitative and qualitative research

Two broader methodologies are mostly distinguished to classify different types of

research studies – quantitative and qualitative research.

18
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

(a) Quantitative research

Quantitative research involves “the collection of primary data from large

numbers of individuals, frequently with the intention of projecting the
results to a wider population” (Bennett, 1996:125). Quantitative research
is normally conducted amongst a representative sample of a target
population with the aim to generalise the research findings to the specific
population (or universum). The emphasis is on numerical measurement
(Smith, 1998:40; Bless and Higson-Smith, 2000:38 and MORI, 2001a)
and subsequent statistical analysis (Smith, 1998:40; Bless and Higson-
Smith, 2000:38). The large sample (number of respondents) is necessary
to analyse the results according to categories within the target population -
that is according to age, gender, exposure to a communication product or
message and so forth. According to Reinard (2001:8) quantitative
research “tends to be explanatory, especially when experiments are
involved, or it attempts to use precise statistical models to achieve
comprehensive understandings of human communication (as in survey
studies and polls of public opinion).” Using quantitative research methods,
researchers often aim to explain communication behaviour by looking at
processes that allow them to predict future behaviour (Reinard, 2001:8).

Reinard (2001:11) argues that there are two major types of quantitative
research - surveys and experiments. Examples of the different types of
quantitative research, in the focus area of communication research is also
provided by Reinard – see table 2.1.

19
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

TABLE 2.1

TYPES OF QUANTITATIVE STUDIES1

TYPES (DESCRIPTION) EXAMPLES

SURVEY METHODS:
Techniques that involve
carefully recorded
observations that provide
quantitative descriptions of
relationships among
variables
Descriptive or • Discovering what sorts of things small-
observational surveys: group communicators say that predict
Direct observation of their becoming group leaders
behaviour by use of some • Identifying the relationship between the
measurement (the number of newspapers a person reads
researcher does not on a regular basis and the amount of
manipulate or change any ear of society the person reports
variables)

(continue …)

1
Reinard, 2001:11

20
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

TYPES (DESCRIPTION) EXAMPLES

Content analysis: • Studying the amount of violence on
A systematic, quantitative children’s television programmes
study of verbally • Inquiring into the amount of newspaper
communicated material space dedicated to stories about a
(articles, speeches, films) women‘s movement
by determining the • Analysing the types of speech defects
frequency of specific ideas, shown by children in samples of
concepts, or terms spontaneous speech
Opinion surveys: • Analysing surveys regarding which
assessments of reports candidate people think won a political
from individuals about debate
topics of interest • Examining whether the public believes
that speech correction therapy should
receive increased funding in public
schools
• Assessing surveys of the favourite
television programmes people watch
EXPERIMENTAL • Studying the impact of the use of
METHODS: evidence by exposing one group to a
A method of studying the speech with evidence and another
effect of variables in group to a speech without evidence
situations where all other • Studying the effect of colour in
influences are held advertising by exposing one group to an
constant. Variables are advertisement with colour printing and
manipulated or introduced another group to an advertisement
by experimenters to see without colour printing
what effect they may have

21
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

The methods of data collection mostly used for quantitative surveys

include:
- face-to-face (or personal) interviews of the interviewer (or
researcher) with the respondent at the respondent’s place of
residence, in the street, at shopping malls or at work
- telephone interviews
- self-completion of a questionnaire by the respondent (either alone
or as individuals in a group) in the presence of the researcher or
through the questionnaire being delivered and collected by the
researcher
- self-completion of a questionnaire by the respondent received and
returned by post
- Internet or online research
(Ward, 1998:149-171; Bennett, 1996:125-133; Smith, 1998:49-55).

Each of these data collection methodologies has certain limitations and

advantages – e.g. in terms of cost, time to collect the data, quality control
and sampling efficiencies. These are not discussed by the researcher.
The limitations and advantages need to be properly investigated and
considered in the context of each survey when a decision has to be taken
regarding the method of data collection.

(b) Qualitative research

Qualitative research, according to Goodyear (1998:177), is often defined

in terms of its relation to quantitative research: “Where quantitative
research measures, and answers questions like ‘how many, how often,
what proportion, what size …?’, qualitative research leads to
understanding and answers questions like ‘why did, how can, in what
way?”

22
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Furthermore, Bennett (1996:133-134) argues that qualitative research

methods can also be used “to uncover new ideas from or hidden feelings
of respondents” and that it can best be achieved by “unstructured
interviews in which respondents can talk freely without too much leading”
from the moderator.

In qualitative research, qualifying words or descriptions are used to record

responses (Bless and Higson-Smith, 2000:37) and observations are
mostly described in “non-numerical terms” (Reinard, 2001:6). In
qualitative research the researcher involves a smaller number of
respondents (Smith, 1998:40) and there is no attempt to generalise about
the population (Bennett, 1996:145). Qualitative research does not place
the same emphasis on classic statistical validity as quantitative research.
Smith (1998:40) explains that, for qualitative research, “validity centers
more on face validity – that is, establishing whether the evidence is
consistent with existing theories and prior knowledge.”

Qualitative research is response and not question orientated. The

response to a question largely determines the following question,
therefore respondents are not interviewed according to a predetermined
set of questions (Smith, 1998:40).

According to Reinard (2001:6) qualitative research studies in the field of

communication “tend to describe or interpret communication exchanges.”
Reinard (2001:7-8) proceeds to explain that these studies attempt to
“describe the human condition by using general views of social action” and
that “researchers who use qualitative methods often try to interpret the
meanings to be found in communication exchanges.”

23
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Qualitative research can be conducted in either an individual or group

setting. There are a number of different approaches to collecting
qualitative results on an individual basis. According to Smith (1998:45),
the depth interview is the most commonly used method in this regard.
Bennett (1996:134) emphasises that depth interviews require the services
of skilled interviewers and refers to Webb (1992) who identifies the
following circumstances where depth interviews will be particularly useful:

- when the issue under investigation is embarrassing, stressful or of

a confidential nature
- when a detailed analysis needs to be conducted of rather complex
situations such as attitudes, beliefs and feelings
- when peer pressure may cause some respondents to conform to
societal norms when in reality they would not
- when the interviewer needs a progressive set of images, such as
buying decision with regard to overseas holidays
- in complex situations when the aim is to explore rather than
measure.

From experience, the researcher can add that depth interviews with
individual respondents are, for various reasons, also often preferred to
qualitative research in a group setting when involving leaders, managers
or opinion-formers.

With regard to collecting qualitative research results on a group basis,

there are also a variety of approaches, but the most widely used is the
group discussion (MORI, 2001b; Smith, 1998:45; Bennett, 1996:136). A
group discussion describes a session involving between six and eight

24
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

individuals (Smith, 1998:45) who are recruited according to specific

criteria.2

The moderator’s role is to manage the discussion flow in the group (Smith,
1998:45). Bennett (1996:138) explains that ideally the moderator should
let the group carry on the conversation “by themselves” and that
“interventions are deemed necessary only to introduce a new topic if it
does not come up spontaneously or to bring the discussion back on track
if participants have strayed into irrelevant areas.” The moderator needs to
manage the discussion flow according to the discussion guide in a
response-orientated approach and probe for response where appropriate.
The moderator should “create a relaxed atmosphere in which respondents
can comment in a constructive, non-defensive way” (Smith, 1998:45) and
facilitate the balanced involvement of different members of the group.
Moderators “should exercise just enough authority to direct and control the
flow of conversation without affecting its content” (Bennett, 1996:138).
Furthermore, the moderator needs to observe non-verbal communication
in the group (Smith, 1998:45).

The terms group discussion and focus group are mostly used
interchangeably, but Smith points out that they are in fact slightly different:
“The group discussion, which has a European pedigree, places the
emphasis on depth understanding. In contrast, the more American style
focus group tends to place more emphasis on – albeit still in a qualitative
mode – measurement and quantification” (Smith, 1998:45). It is of critical
importance to properly understand the need of the research client in this
regard.

2
The criteria will depend on the nature of the study. In communication research basic
socio-demographic variables are commonly used – e.g. gender, age, level of socio-
economic development and language preference.

25
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

2.3.2.2 Basic and applied research

Based on the reasons why research is conducted, the two types of research
distinguished are basic research and applied research. Whether the aim of the
research is basic or applied does not affect the way in which the research is
conducted (Bless and Higson-Smith, 2000:39) – the methods of inquiry are
identical (Reinard, 2001:4). Suchman (cited by Philips, 1985:534) and Kidder
and Judd (1986:396) also emphasise that the significant difference between
basic research and applied research is one of purpose and not of method.

(a) Basic research

Basic research applies when the researcher seeks to “contribute to human

knowledge and understanding relating to a specific phenomenon” (Bless and
Higson-Smith, 2000:38). Neuman (1997:21) argues that “basic research
advances fundamental knowledge about the social world.” Neuman furthermore
explains that basic research “focuses on refuting or supporting theories that
explain how the social world operates, what makes things happen, why social
relations are a certain way, and why society changes” and that “basic research is
the source of most new scientific ideas and ways of thinking about the world”
(Neuman, 1997:21). Reinard (2001:4) explains that basic research is conducted
“to learn about relationships among variables” and according to Vickery (cited by
Powell, 1997:2) basic research “is concerned with elucidating concepts and their
relations, hypotheses and theories.” This aim is normally achieved by “gathering
more facts and information which enables existing theories to be challenged and
new ones to be developed” (Bless and Higson-Smith 2000:38).

Basic research is research conducted “regardless of any immediate commercial

product or service” (Reinard, 2001:4). As Bless and Higson-Smith (2000:38) put
it, the “actual utility or application of this newly acquired knowledge is of little

26
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

concern to the researcher.” According to Reinard (2001:34) and Powell (1997:2)

most research usually referred to as ‘pure’ scientific research is actually basic
research.

(b) Applied research

Applied research is conducted if the researcher’s motivation is to assist in solving

a particular problem (Reinard, 2001:4; Bless and Higson-Smith, 2000:38;
Philips, 1985:534; Neuman, 1997:22 and Powell, 1997:2) or to develop a
product (Reinard, 2001:4). According to Neuman (1997:22) theory is “less
central” to applied researchers than “seeking a solution to a specific problem”
and its main strength, is its “immediate practical use.”

Although Powell (1997:2) mentions that applied research is occasionally referred

to as action research, Neuman (1997:23-28) remarks that researchers use
“several types of applied research,” and distinguishes three types of applied
research: action research, social impact assessment and evaluation research.

According to Neuman (1997:23) “action research is applied research that treats

knowledge as a form of power and abolishes the line between research and
social action.” Isaac and Michael (cited by Powell, 1997:45) state that the
purpose of action research is “to develop new skills or new approaches and to
solve problems.” Action research is characterised as “practical, orderly, flexible
and adaptive, and empirical to a degree, but weak in internal and external
validity” (Isaac and Michael, cited by Powell, 1997:45). Neuman (1997:23)
explains that there are “several types of action research” and that “most share
the following common characteristics:
• those who are being studied participate in the research process
• research incorporates ordinary or popular knowledge
• research focuses on power with a goal on empowerment

27
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

• research seeks to raise conscious or increase awareness, and

• research is tied directly to political action.

Social impact assessment, according to Neuman (1997:24) “may be part of a

larger environmental impact statement required by government agencies. Its
purpose is to estimate the likely consequences of a planned change. Such an
assessment can be used for planning and making choices among alternative
policies.” Researchers conducting social impact assessments “examine many
outcomes and often work in an interdisciplinary research team” (Neuman,
1997:24). Neuman furthermore notes that “social impact studies often include a
cost-benefit analysis” in which the researcher “estimates the future costs and
benefits of one or several proposed actions.”

Evaluation research is a widely used type of applied research (Neuman,

1997:25) and has as its primary goal “not the discovery of knowledge but rather a
testing of the application of knowledge within a specific programme or project”
(Powell, 1997:45). Neuman (1997:25) emphasises that “ethical and political
conflicts often arise in evaluation research because people have opposing
interests in the findings” and Powell (1997:45) notes that “evaluative researchers
must be concerned with threats to validity, such as intervening variables,
measurement techniques and operational definitions.” The two general types of
evaluation research are summative evaluation and formative evaluation.
Summative evaluations look at final programme outcomes (Neuman, 1997:25).
A summative or outcome evaluation “tends to be quantitative in nature and often
is used as the basis for deciding whether a programme will be continued”
(Powell, 1997:46). Formative or process evaluation is “built-in monitoring or
continuous feedback on a programme” (Neuman, 1997:25) and “examines how
well the programme is working” (Powell, 1997:46). According to Powell
(1997:46) formative evaluation is often more qualitative and it is typically used for
“revising and improving programmes.”

28
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Classifying research projects in this way is perceived as not being very useful in
practice. Bless and Higson-Smith (2000:39) argue that no study is “only purely
basic or purely applied,” and Reinard (2001:4) explains that “last year’s basic
research may be today’s source of new products.” Powell (1997:2) also holds
the opinion that basic and applied research are “not necessarily dichotomous”
and that “in spite of the fact that basic and applied research have tended to be
conducted in isolation from one another.”

2.3.2.3 Different objectives of social research

A third way of classifying types of research is based on the demands of the

research question, that is in terms of the research objectives. Bless and Higson-
Smith (2000:37-44) distinguish four types of research on the basis of this
classification: exploratory, descriptive, correlational and explanatory.

(a) Exploratory research

In cases where very little is known about the research topic, one speaks of
exploratory research (Bless and Higson-Smith, 2000:37). Powell (1997:58-59)
explains that exploratory research “can increase the researcher’s familiarity with
the phenomenon in question, can help to clarify concepts, can be used to
establish priorities for future research, can identify new problems and … can be
used to gather information with practical applications.” According to Neuman
(1997:19) exploratory researchers are “creative, open minded, and flexible;
adopt an investigative stance; and explore all sources of information.
Researchers ask creative questions and take advantage of serendipity, those
unexpected or chance factors that have large implications.”

Both Powell (1997:58) and Neuman (1997:19) remarks that exploratory

researchers frequently conduct qualitative research. Powell (1997:59), speaking

29
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

of exploratory research in general, emphasises that “it is important to remember

that exploratory studies merely suggest insights or hypotheses; they cannot test
them,” and Smith (1998:38) remarks that a “typical outcome” from exploratory
research would be “the generation of a number of hypotheses that could be
taken forward for quantitative testing at a later stage of the project.”

(b) Descriptive research

Descriptive research “presents a picture of the specific details of a situation,

social setting or relationship” (Neuman, 1997:20). Bless and Higson-Smith
(2000:41) and Neuman (1997:20) are of the opinion that descriptive and
exploratory research have some similarities. Neuman (1997:20) remarks that
descriptive and exploratory research “blur together in practice” and explains that
“in descriptive research, the researcher begins with a well-defined subject and
conducts research to describe it accurately.” Descriptive research focuses on
“how”, “who”, “what”, “when” and “where” questions (Smith, 1998:38 and
Neuman, 1997:20) and “provides a solid platform for helping to understand
currents, and possibly predict future behaviour” (Smith, 1998:38).

(c) Correlational research

When the research question requires an understanding of the relationship

between variables, the research is called correlational research (Bless and
Higson-Smith, 2000:37). The task of determining a casual relationship is a
complex and difficult one. Bless and Higson-Smith (2000:43) explain that “it is
based on systematic comparison, manipulation and control of variables.”
Correlational research is “not only useful when no clear causal relationship
exists, but also allows for an estimation of the strength of the relationship
between two variables even when one variable is influenced by many others”
(Bless and Higson-Smith, 2000:43).

30
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

(d) Explanatory research

“When the research question demands that the researcher explains the
relationship between variables and demonstrates that change in one variable
causes change in another variable, the research is called explanatory research”
(Bless and Higson-Smith, 2000:37). According to Neuman (1997:20) the desire
to know why things are the way they are, to explain, is the purpose of
explanatory research. Neuman (1997:20-21) explains that explanatory research
“builds on exploratory and descriptive research and to identify the reason why
something occurs.” Bless and Higson-Smith (2000:43) state that explanatory
research is often not feasible: “This is the case when it is not possible to
manipulate the suspected independent variable or to assess the time-order of
variables.”

2.3.3 Summary

Different research methodologies are applied (or applied in combination)

depending on the reasons why the research is conducted, the demands of the
research question, the target group for the research, available funding, time-
scales and the competency and capacity of researchers.

2.4 THE RESEARCH PROCESS

2.4.1 Introduction

In paragraph 2.2 the researcher, with reference to various authors, mentions that
research is a “systematic effort”, a “systematic process” and the “systematic and
objective collection, analysis and evaluation of information.” In paragraph 2.5.3,
reference is made to Reinard (2001:12) who remarked that “productive research

31
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

follows steps that carry out some sort of design.” This implies that every
research project invariably requires careful and appropriate planning and
execution by the researcher.

No matter how unique any research problem or project, there are a number of
steps which are common to the process relevant to most research projects. As
the word process implies, Puth (1996:80) explains, “(marketing) research
involves a series of steps or phases which cannot be viewed in isolation, but
which should be seen and dealt with as an integrated whole. This integrated
evolvement of steps which are followed when planning and executing a research
project is known as the research process.”

Although not all the steps are applicable to all types of research, sensitivity and
application of relevant steps in the research process by the researcher will
enhance the success and quality of the research project – it will assist the
researcher in his/her initiative to do “the right thing” in “the right way” (see
paragraph 2.1).

Different sources of research literature distinguish many different steps and

permutations of steps. The researcher uses the steps as identified and
developed by Puth (1996:80-96) through combining and synthesising an
extensive variety of relevant documentary resources. The different steps in the
research process are referred to shortly, but not discussed in any detail.

2.4.2 Steps in the research process

2.4.2.1 Identifying and formulating the problem

Bless and Higson-Smith (2000:15) express a sentiment shared by many other

researchers that “selecting a research problem is a delicate task.” Puth

32
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

(1996:82) mentions that the American Marketing Association (AMA) came to the
conclusion in 1968 that “if any step in a research project can be said to be more
important than the others, then problem definition is that step.”

The step in the research process referred to by the AMA as ‘problem definition’,
is broken up into two distinguishable phases by Puth (1996:82). The first phase
entails the identification of the problem whilst the second phase is the refinement
of the problem to a level where it can be clearly formulated in order to provide
direction and guidance to the research process. If the research problem is well
formulated and the research objectives precisely defined, the likelihood of
designing a research study that will provide the necessary information in an
efficient manner is greatly increased. Problem identification and problem
formulation should result in “a precise statement of the objectives of the research
to be conducted and a set of research questions” (Puth, 1996:82).

Reinard (2001:32-35) suggests the following five criteria to formulate sound and
useful problem statements:
• problem statements must be stated unambiguously, usually as questions
• except for simple exploratory studies, problem statements must include at
least two variables
• problem statements must be testable
• problem statements must not advance personal value judgements
• problem statements must be clear grammatical statements.

2.4.2.2 Deciding on what kind of data is required

After identifying and formulating the problem, the researcher will have a good
idea of the nature of the data required. According to Puth (1996:84) the data
relating to the research objectives may be anyone or more of the following:

33
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

• facts - e.g. demographic profile

• levels of awareness - e.g. awareness of an announcement by
government regarding initiatives to create job opportunities
• opinions and attitudes - e.g. respondents’ opinions and attitudes on
whether government’s initiatives to create jobs will be successful
• preferences - e.g. the channel(s) of communication preferred (closely
related to opinions and attitudes)
• motives or predispositions - reasons why people act or think as they do
• behaviour (that can be seen as the result of attitude, preference and
motivation).

On their website, MORI (2001c) also adds the aspect of knowledge – that is
assessing what the respondents know (or think they know, or claim to know!).

2.4.2.3 Exploring secondary data sources

Puth (1996:86) argues that as every research project is a search for information
on some topic, researchers can be more confident of the quality and
appropriateness of their information if they tap all the relevant resources. “Often
there is a wealth of information and data on the research problem already
collected by others, in which case it may not be cost-effective or necessary to
conduct a whole new research project in order to answer the research question.
In many cases existing secondary data may be sufficiently relevant and
comprehensive to answer at least a certain part of the overarching research
question” (Puth, 1996:86).

An exploration of secondary data resources can begin with a search of published

data, identification of unpublished data that is relevant and interviewing
knowledgeable or well-informed people on the topic or problem area. It is

34
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

essential to explore all possibilities of secondary data sources before proceeding

with the remaining steps of the research process.

Reinard (2001:76) remarks that “some inexperienced researchers believe they

can ignore past work and use entirely new ideas and methods”. Reinard
(2001:76) proceeds by referring to Stanovich who calls this misguided approach
the ‘Einstein syndrome’ since “researchers who suffer from it fail to connect their
‘sudden breakthroughs’ with lessons from others. By discarding previous
lessons as irrelevant, they fail to learn from the lessons from others.”

2.4.2.4 Revising and fine-tuning the research question

After exploring the secondary data sources, the researcher needs to fine-tune the
research question. According to Puth (1996:86), this is the stage at which “a
clearer picture of the problem starts to emerge and where the project begins to
crystalise in one of two ways:
• it is apparent that the question has been answered and the research
process has been completed
• the original question has been modified in some way by the gathered
information.”

Puth (1996:86-87) furthermore identifies five other problem-related activities that

should be considered to effectively complete the fine-tuning of the original
question:
• examine if the concepts and constructs to be used in the investigation are
defined satisfactorily
• review the investigative questions to break them down into more specific
levels of questions
• if hypotheses are used, they must be relevant to the refined research
problem

35
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

• determine what evidence needs to be collected to answer the various

questions and hypotheses
• set the boundaries or limits of the project by stating what is part of the
research problem and what is not.

2.4.2.5 Designing the research study

Mouton (2001:55) explains that a research design is “a plan or blueprint of how

you intend conducting the research.” Mouton (2001:56) argues that researchers
often confuse ‘research design’ and ‘research methodology’ and summarises the
differences between these two concepts – see table 2.2.

TABLE 2.2

DIFFERENCES BETWEEN RESEARCH DESIGN AND RESEARCH

METHODOLOGY – A SUMMARY3

RESEARCH DESIGN RESEARCH METHODOLOGY

Focuses on the end product: Focuses on the research process and
What kind of study is being planned the kind of tools and procedures to be
and what kind of result is aimed at? used
Point of departure = research problem Point of departure = specific tasks
or question (data collection or sampling) at hand
Focuses on the logic of research: Focuses on the individual (not linear)
What kind of evidence is required to steps in the research process and the
address the research question most ‘objective’ (unbiased) procedures
adequately? to be employed

3
Mouton, 2001:56

36
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Puth (1996:87) is of the opinion that “selecting an appropriate research design is

often complicated by the availability of a large variety of methods, techniques,
procedures and ever-more-sophisticated computer programming and
technology.” Also emphasising the reality that the design of the research study is
one of the most challenging steps in the research process, is the viewpoint of
Smith (1998:29) that researchers will seldomly be able to pursue their ‘ideal’
design and that it is a process of compromise. Smith (1998:29) explains that
“decisions have to be made about what degree of precision is needed and how
much depth of understanding is required. This trade-off also needs to be
balanced against the time and budget available.” Further, Smith explains,
“market researchers must take into account the practicality of different
approaches whilst ensuring the study is ethical” and also complies with the codes
of conduct in the research industry.

Mouton (2001:57) presents a broad classification of the main research design

types according to the kind of questions the design types are able to answer –
see figure 2.1.

37
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Figure 2.1: A typology of research design types4

TYPES OF STUDY

Empirical studies Non-empirical studies

(Philosophical analysis, conceptual analysis,

theory building, literature reviews)

Using primary data

(Surveys,
experiments, case Analysing existing data
studies, programme
evaluation,
ethnographic studies)

Text data Numeric data

(Discourse analysis, (Secondary

content analysis, data analysis,
textual criticism, statistical
historical studies) modeling)

2.4.2.6 Determining the sample

According to Collins (1998:69) “almost all market research studies use sampling
– the attempt to learn about some large group, a population, by looking at only a
small part of it, a sample.”

4
Mouton, 2001:57

38
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

The “population” referred to by Collins is not necessarily the total population of a

country or area, but the totality of the target group (or universum) from which the
sample needs to be drawn. As a first step in the sampling process the target
population needs to be identified. Thereafter the researcher needs to determine
the sample characteristics and determine the sample size.

Two broader sampling methods can be distinguished, namely probability and

non-probability sampling. Martins (1996d:253) explains that “a probability sample
is one in which every element has a known non-zero probability of being
selected. It is unnecessary for all elements to have an equal chance of being
selected, but each element must have a chance and that chance must be known
so that the sampling results can be applied to the universe. Non-probability
samples rely on the judgement of the researcher and are only as representative
as the researcher’s luck and skill permit. In non-probability sampling there is no
way of estimating the probability that any element will be included in the sample,
and therefore there is no method of finding out whether the sample is
representative or not”. It is important to note that “the most important criterion” of
a sample, according to Puth (1996:87) is that it will be “totally representative of
the population relevant to the solving of the management problem and the
ensuing research questions.”

2.4.2.7 Allocating funds and resources

Depending on the nature and scope of a research project, substantial financial

and human resources may be necessary. The researcher needs to do
appropriate planning and allocate resources timeously in order to avoid a
situation where a project has to be terminated due to a lack of resources.

Puth (1996:88) explains that “although data collection does require substantial
resources, it might not always be as big a part of the budget as clients or

39
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

researchers would expect. Employee salaries, training and travel, and other
miscellaneous expenses are incurred during data collection, but this phase of the
project often takes no more than a third of the research budget. The geographic
scope and number of respondents naturally affect the cost, but much of the cost
is relatively independent of the extent and size of the data gathering exercise.”
An interesting and useful guideline suggested by Puth (1996:88) is that project
planning, data collection and analysis and, lastly, interpretation and reporting
each have a share more or less equal in the budget.

2.4.2.8 Writing and presenting the research proposal

A research proposal is mostly developed and fine-tuned concurrently with the

exploring and planning phases of the research project. The research proposal
would therefore incorporate the decisions and choices made by the researcher in
the preliminary stages of the project.

The most important purpose of the research proposal is to ensure that all parties
concerned understand the project’s purpose and the proposed methods of
research. Time limits and budgets are also identified and justified in most
research proposals. Various responsibilities and obligations are clarified.

According to Puth (1996:89) “every proposal should contain two basic sections,
namely the problem statement and a statement of what will be done and how it
will be done. In its varied forms the research proposal can include any number of
the following elements: executive summary, problem statement, research
objectives, literature review, importance and benefits of the study, research
design, data analysis, nature and form of results, qualification of researchers,
budget, time schedule, facilities and special resources, project management,
bibliography and appendices.”

40
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

2.4.2.9 Conducting a pilot test

Piloting is the last stage in the design of a research project (including design of
the questionnaire or discussion guide) before the survey goes into the field for
data collection (Miller and Read, 1998:380).

Puth (1996:89) explains that the primary purpose of a pilot test is two-fold: to
detect weaknesses in design and instrumentation and to provide a sound base
for determining and refining the sample. During the pilot test respondents are
drawn from the universum or target population and the procedures and protocols
of the research project are simulated. The number of respondents involved in a
pilot test would depend on primarily the research methodology and the
characteristics of the target population, but need not be statistically selected.

2.4.2.10 Collecting primary data

According to Bulmer (2000b:205), the “most critical phase in social research is

that during which data are actually collected.”

The different data collection methods (e.g. questionnaires or transcribed

recordings of focus group discussions) will have different implications for data
collection. Each method of data collection has specific advantages and
disadvantages, qualifying it as a better or a less-preferable option than other
methods for the collection of certain types of data. Puth (1996:90) mentions that
“although a combination of methods can be considered in certain circumstances,
it is often not done for reasons of cost.” The researcher then needs to decide
invariably on the method that will yield the most satisfactory range of reliable data
as cost-effectively and as quickly as possible.

41
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

In order to provide data in a form that can be used by the researcher for analysis
and interpretation, it needs to be edited to “identify and isolate omissions and
spoilt responses. In the case of survey methods editing is essential to reduce
recording errors, to improve legibility, and to identify and filter unclear and
inappropriate responses” (Puth, 1996:90).

2.4.2.11 Analysing and interpreting the data

Data analysis involves reducing the accumulated data to a manageable size to

allow summarising, comparing, syntheses and applying statistical techniques in
order to interpret the results in relation to the research problem. Data analysis
can be either very basic (e.g. one-way frequency distributions or cross-
tabulation), involve different methodologies of significance testing (e.g. analysis
of variance or the Mann-Whitney test) or even done by applying multivariate
statistical techniques (e.g. discriminant and cluster analysis) (Martins, 1996e:305
and 315; Loubser, 1996b:336 and 339; Wegner, 1996:356-363).

Mouton (2001:109) explains that interpretation involves the synthesis of data into
larger coherent wholes. Observations or data are interpreted and explained by
“formulating hypotheses or theories that account for observed patterns and
trends in the data. Interpretation means relating one’s results and findings to
existing theoretical frameworks or models, and showing whether these are
supported or falsified by the new interpretation. Interpretation also means taking
into account rival explanations or interpretations of one’s data and showing what
levels of support the data provide for the preferred interpretation.”

In order to interpret results correctly, the researcher needs to be familiar with the
method of the research and the limitations of the results (Van Wyk, 1996:396).
The pitfalls awaiting the researcher in the interpretation of results include the
following:

42
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

• drawing inferences from the data that are not supported by the data
• biased intepretation of the data through selectivity
• overgeneralisation
• confusing correlation with causation
(Van Wyk, 1996:396 and Mouton, 2001:110).

Adding to the pitfalls already mentioned, Van Wyk (1996:397) emphasises “there
may be more information hidden in the data than the researcher cares to or is
able to bring to light” and that “it requires experience, disciplined thinking and
familiarity with the research method to let the results say what they are able to
say.”

2.4.2.12 Reporting the results

Preparing the research report and communicating the research findings and
recommendations to the client are the final steps in the research process. Van
Wyk (1996:398) argues that “the report is the culmination of the whole research
project” and quotes Churchill who expressed this sentiment: “Regardless of the
sophistication displayed in the other portions of the research process, the project
is a failure if the research report fails.”

The ultimate objective with the report is “to enable the client to make an informed
and scientifically verified decision to solve the original problem that prompted the
undertaking of research in the first place” (Puth, 1996:90). Various authors
emphasise that research reports will be quite different in terms of style and
organisation depending on the aim and objectives of the research project and the
target audience for the report (e.g. Van Wyk, 1996:398-402; Puth, 1996:90;
Bless and Higson-Smith, 2000:141).

43
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Puth (1996:91) remarks that “the items to be included in a research report are
essentially the same as those identified in the discussion of the research
proposal.” But, taking into account the target audience for the research report,
Bless and Higson-Smith (2000:141) write as follows:

The most detailed, complete and scientific report for research-funding institutions
and archives will present all the different steps of research in detail. A report
written to be published in a scientific journal will have to show a high level of
scientific quality condensed into a few pages. A report written for an agency
particularly interested in the conclusions and practical consequences will cut
short the technical aspects of the research and emphasise the discussion of the
findings. A report to be understood by the average educated readership of a
magazine will present the findings in more general terms and will avoid scientific
vocabulary. In other words, these different reports will stress one or the other
aspect of the most complete research report.

Clients will often also expect the researcher to make a personal presentation of
the findings and recommendations. Marbeau (1998:520) is of the opinion that
the challenge of results presentation lies in achieving “speed and clarity without
stripping out any important substance from the findings. The answer is
conciseness, i.e. being short yet complete.” The same author also regards it as
important for the presenter to be modest, and to “present the results and the
answers rather than the research and the researcher. Also important is honesty,
to separate the reliable facts from their hypothetical interpretation” (Marbeau,
1998:520).

44
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

2.5 CHALLENGES OF COMMUNICATION RESEARCH

Already from the preceding paragraphs, it is clear that researchers conducting

communication research are exposed to various problems and challenges.
Awareness of such challenges is essential for conducting communication
research of professional quality.

Reinard (2001:6-14) identifies the following six challenges of communication

research:
• the challenge of breadth and focus
• the multiple methods challenge
• the scholarly rigour challenge
• the personal challenge (or what do I need to do to study communication
research methods successfully?)
• the ethical challenge
• the structure of the field challenge.

The researcher shortly discusses the challenges of communication research

according to Reinard’s categories of challenges, incorporating contributions from
other authors. A few of the other challenges facing the communication
researcher are also mentioned.

2.5.1 The challenge of breadth and focus

Reinard (2001:6) is of the opinion that although “the number of communication

applications can seem enormous, there is a rational order to it.” Making this
statement, he refers to the work done by McBath and Jeffrey (1978) to identify
the professional areas in communication on behalf of the Speech Communication
Association and the National Center for Educational Statistics (NCES). These
two organisations were trying to organise information about careers in various

45
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

fields. The list of communication specialities used by the NCES is provided in

table 2.3. The left column of the table shows the official taxonomy with an
emphasis on the career areas of scholars. Reinard added a couple of areas to
those originally listed by McBath and Jeffrey - e.g. conflict management,
journalism, radio and television, public relations and health communication. In
the column to the right of table 2.3 a description is provided of the kinds of
research issues that are normally addressed in each of the areas.

TABLE 2.3

COMMUNICATION SPECIALITIES5

COMMUNICATION TAXONOMY DESCRIPTION

1. The Broad Areas of Mass

Media Communication
Advertising The study of mass media methods of influence
to promote a product, service or cause
Communication Technology The study of the mechanisms and technologies
of mass media
Communication Policy The study of public policy and regulation of
mass media communication and freedom of
speech
Film as Communication The role of popular and technical cinema in
society
Journalism The study of the methods of reporting and
organising news for presentation in print media

5
Reinard, 2001:7-8

46
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Public Relations The study of methods of managing publicity and

press relations for an organisation, person or
cause
Radio The study of the methods and uses of radio
Television The study of the methods and uses of televised
communication
2. Specific Areas of Speech
Communication Research
Code Systems The study of the uses of verbal and non-verbal
symbols and signs in human communication
Intercultural Communication The study of communication among individuals
of different cultural backgrounds
Interpersonal Communication The study of communication interactions
occurring in person-to-person and small group
situations
• Conflict Management The study of the role of communication in the
creation and control of conflict
• Family Communication The study of communication transactions within
the constraints of families of all sorts
Organisational Communication The study of interrelated behaviours,
technologies, and systems functioning within an
organisation
• Health Communication The study of communication issues among
participants involved in medical and health
systems
Oral Interpretation The study of literature through performance
involving the development of skilled verbal and
non-verbal expression based on critical analysis
of written texts (aesthetics of literature in
performance, criticism of literature in
performance, group performance, oral traditions)

47
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Pragmatic Communication The study and practice of communication, the

object of which is to influence or facilitate
decision making
• Argumentation The study of reason-giving behaviour
• Debate The study of decision making in which
adversaries present arguments for decision by a
third party
• Discussion and Conference The study of methods of decision making in
(including Group Decision which participants strive to discuss, explore and
Making) make decisions on issues
• Parliamentary Procedure The study of the means used to handle
deliberation in large legislative bodies through
the use of formal rules and procedures to
regulate debate and discussion
• Persuasion The study of the methods used to influence the
choices made by others
• Communication and the The study of communication issues involved in
Law the legal system and the practice of law
Public Address The study of speakers and speeches, including
the historical and social context of platforms,
campaigns, and movements
Rhetorical and Communication The study of the principles that account for
Theory human communicative experiences and
behaviour
Communication Education The study of communication in pedagogical
contexts (communication development, oral
communication skills, instructional
communication)

48
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Speech and Hearing Science The study of the physiology and acoustical
aspects of speech and hearing (biological
aspects of speech and hearing, phonological
aspects of speech and hearing, physiological
aspects of speech and hearing)

The researcher fully agrees with Reinard’s (2001:6) sentiment that “each area is
broad enough to promote many interesting studies.” The breadth and extensive
focus of communication research clearly poses various challenges to the
researcher, and the importance of having a broad knowledge should be
emphasised.

2.5.2 The multiple methods challenge

Reinard (2001:6) explains that qualitative methods are mostly used in studying
literature whilst the historical method is employed in history and the experiment
holds a prominent position in psychology. Contrary to this, the communication
researcher uses all of these and other methods to answer questions. Research
projects in communication tend to rely mostly on quantitative and qualitative
methods – see paragraph 2.3.2.1. The research question or objectives guides
the selection of methods, not the other way around (Bulmer, 2000a:10; Reinard,
2001:8; Smith, 1998:40).

Smith (1998:39) remarks that ”the debate that is often conducted about the
merits of small scale, flexible qualitative, and larger scale, structured quantitative
research, has been largely unhelpful in the sense that they are more ‘mutual
friends’ than ‘mutually exclusive foes’.” To a growing extent it is suggested that
different methodologies should be used in combination. Smith (1998:40) strongly
argues the case that “a good research design invariably involves adopting an
eclectic approach, mixing together the best combination of methodologies to deal

49
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

with the research objectives.” Sonnenwald and Iivonen (1999:430-431) conclude

that “research in information studies increasingly combines multiple methods to
research human information behaviour because doing so can provide a more
holistic and comprehensive view of information behaviour, increase the validity of
research results through data and methodological triangulation, or both.”

The communication researcher needs to have appropriate knowledge of different

research methods and needs to know when, why and how to apply different
methods. Furthermore, the researcher needs to understand when and how to
apply more than one method during work on a communication project in order to
properly address the aim and objectives of the research project.

2.5.3 The scholarly rigour challenge

In order for research to meet standards of excellence, communication

researchers must conduct research with recognition of five key challenges
identified after Tuckman by Reinard (2001:12):
• Research is systematic. Productive research follows steps that carry out
some sort of design. Researchers ask questions and implicitly agree in
advance to seek for answers by examining pertinent information.
• Research is data driven. If data cannot be collected, or if we are unwilling
to alter our opinions, the issue is not suitable for research.
• Research is a sound argument. Research arguments reason from
research data and information to draw conclusions. Thus, arguments in
this context are defined as claims advanced on the basis of reasoning
from evidence. Sound reasoning is vital for effective research. Logic and
the methods to evaluate arguments are valuable tools to judge research.
• Research is capable of replication. If research methods are so vaguely
described that it is impossible to repeat the procedures in a study, the
worth of the entire research project is questioned. Regardless of whether

50
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

replications actually are completed, the ability to replicate studies is

essential for any piece of sound research.
• Research is partial. Research findings are partial because we may
discover new relationships involving other variables that make us modify
or qualify the conclusions we have found. Thus, communication
researchers do not claim to have discovered ‘The Truth’ for all time.
Instead, they advance tentative - but meaningful - insights for
communication phenomena.

2.5.4 The personal challenge

This challenge relates to the question of what a person needs to do to study

communication research methods successfully. According to Reinard (2001:13-
14) the following five aspects apply:
• the need to think in an orderly way – to train our minds to separate the
relevant from the irrelevant, the observable from the unobservable, and
the complete from the incomplete
• the need to write clearly – crisply, clearly, precise, structured and to the
point
• the need to set aside personal prejudices in the light of data – be willing to
let the data decide our conclusions, even if we do not like them
• the need to stay organised and follow instructions – research requires
carefully following protocols and methods and to fight the urge to leave out
steps, to take shortcuts or to ignore instructions
• the need to know the reasons for studying research methods – e.g. to
learn to think rigorously and critically, to find answers to questions about
communication, to acquire survival skills to help read and use the field’s
literature and to learn how to sort through past research for answers to
research questions.

51
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

2.5.5 The ethical challenge

Every decision made in communication – e.g. which methodology to apply for a

project – is not merely a practical one, but also an ethical one. Reinard
(2001:14) remarks that “research is judged not only by the rigor of procedures
and the results obtained but (also) by the ethics of the researchers.”

In the literature on social research, the importance of specifically the ethical

issues of voluntary participation of respondents, privacy, anonymity and
confidentiality is emphasised (e.g. Bless and Higson-Smith, 2000:100-101;
Shalofsky, 1998:1111-1113; Groenewald, 1981:97-98). The reality is that ethical
issues in communication research are much broader than the few mentioned
above that only relate to a specific aspect of data collection. For example,
Reinard (2001:14) refers to a situation in the United Kingdom where the
archbishop of York once challenged British scientists to consider the ethical
consequences of their research by urging them to ask “What applications will be
made of my research?” before they undertake their studies.

Over the period of the last few decades, many research organisations have
developed formal codes of conduct to guide practitioners and researchers. The
first code was published in 1948 by the European Society for Opinion and
Marketing Research (ESOMAR) (ESOMAR, 2000b).

In South Africa initiatives of the Southern African Marketing Research

Association (SAMRA) to develop a code of conduct for this industry in the
country go as far back as 25 October 1963. On that date a first sub-committee
was appointed to “investigate the ethical codes of conduct existing in market
research practices overseas.” In October 1967 the first SAMRA Code of
Conduct was ratified. Following various revisions, the current SAMRA Code of

52
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Conduct includes guidelines regarding the following issues (SAMRA, 2001:60-

67):
• responsibilities to the reader of a report
• obligations to the client/sponsor of a survey
• responsibility to informants
• responsibilities to and of the research practitioner
• responsibilities to the general public.

A document that creates food for thought to the researcher sensitive to the
importance of ethics, is the document compiled by Paul Reynolds in 1979.
Reynolds compiled an extensive list of ethical issues from documentation from
24 organisations doing research in the social sciences (Reinard, 2001:14). The
list involves a total of 78 guidelines according to the following broader structure,
and is attached as an Annexure:
• general issues related to the code of ethics
• decision to conduct the research
• conduct of the research
• effects on and relationships with the participants’ informed consent
- general
- provision of information
- voluntary consent
• protection of rights and welfare of participants
- general issues
- deception
- confidentiality and anonymity
- benefits to participants
- effects on aggregates or communities
• interpretations and reporting of the results of the research

53
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

2.5.6 The structure of the field challenge

Reinard (2001:14) correctly remarks that “communication research has been

promoted by many organisations whose members often cross the barriers
created by the organisation of different schools” and that “it is helpful to know
how the diverse and major organisations in our field showcase our research.”

The researcher is of the opinion that communication researchers need to make a

specific effort to enhance their perspective and the quality of their work by
seeking to engage and interact with other researchers in the field of
communication research through outreach to professional organisations in their
country as well as abroad, to private sector research companies, the advertising
and public relations industry and the academic environment.

2.5.7 Other challenges

A few of the many other challenges facing the communication researcher are:
• measurement in a cross-cultural environment (McGorry, 2000:74-79)
• challenges in underdeveloped and developing communities - e.g.
availability of reliable statistics, language, cultural and custom-related
issues, identification, training and management of interviewers (e.g.
Bulmer and Warwick, 2000:38; Gil and Omaboe, 2000:42; Hershfield et
al., 2000:241; Loubser, 1996a:236-248)
• public education on how survey statistics and differences are generated,
and how to use them (Cooper, 1998:1024)
• choosing the right research company to conduct a research project
(MORI, 2001c)
• bridging the communication gap that often exists between researchers
and creatives (Hansen, 1998:655).

54
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

When conducting communication research for government, researchers need to

be alert of all the potential challenges, and implement procedures and processes
to ensure that the research conducted is of unquestionable quality.

2.6 THE USE OF COMMUNICATION RESEARCH BY

GOVERNMENTS

According to Faure (1995:11) previous initiatives for development in “third world”

countries failed due to “lack of understanding by professionals or change agents
for the real needs of ‘third world’ communities. An overarching problem was that
development programmes in the ‘third world’ were characterised by a strong
deterministical (unilateral Western) paternalistic ‘top-down’ approach.”

Interesting though, is information provided by Khoury (1989:77-79) that “even in

ages and societies where government was of despotic nature, most of the
illustrious rulers retained by History, have shown in one way or another a deep
concern for public opinion.” Examples provided by Khoury to prove his point,
include that of H Al Rashid and Cathrine II from Russia. Al Rashid, one of the
most famous Arab Khalifs6, as depicted in the One thousand and one nights tales
of the thirteenth century, was so eager to know what the people thought of his
Khalifat that he left the Palace, and went into the streets of Bagdad asking the
people their opinion. Cathrine II, the Great of Russia, devised a way for taking
into account the public’s reactions in the process of her decision-making. Before
taking an important stand or approving a new law, she used to spread rumours
about it in the streets of Moscow. Thereafter she asked for feedback reports on
the people’s reactions and considered them in her actions.

6
A Khalif in Islam has two roles – he is the head of the state and the head of the
religious authority

55
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Two former heads of state in the United States also remarked about the
importance of public opinion. Worcester (1999:1) quotes Abraham Lincoln as
saying: “Public opinion is everything.” According to Cooper (1998:1015) former
US President Clinton is said to have remarked just prior to his election that “the
most important people in the United States today are those sitting in focus
groups.”

Cooper (1998:1016) argues that “market research is a powerful and well-

established tool for the development and maintenance of any democratic
society.” The term democracy originates from the ancient Greek, demos
(people) and kratos (strength or power). “In essence”, Cooper (1998:1016)
explains, “it means that the strength of a society rests with the people, and that a
society is strong when the people or their elected representatives directly
exercise their power.”

In the first edition, in 1988, of Fundamentals of social research methods: an

African perspective, Bless and Higson-Smith (2000:iii) remark that “it was
observed that many African governments, non-government and private sector
organisations were beginning to attach greater value to social research and the
information it provides.”

Talking about the process of transforming government communication in a

democratic South Africa at the Conference of Government Communicators,
Thabo Mbeki (1995b:1) made clear his opinion about the importance of South
African citizens’ contributions:

The road to an informed and active citizenry should be defined by the citizens
themselves. We must strive to ensure that each individual, whatever his or her
station in life, plays a meaningful role in decision-making and in governance.

56
 U
Unniivveerrssiittyy ooff PPrreettoorriiaa eettdd –– SSttrryyddoom
mMM SS

Mbeki (1995b:2) furthermore remarked that South Africa “cannot afford a

situation where the majority of our people are mere consumers of information
and opinion whose content is determined by one sector of society” and that “the
people out there are crying out to hear and be heard.”

It is encouraging to note that governments increasingly realise that

communication research can be used to enhance the effectiveness of their
communication and the dissemination of government information. In order to
conduct communication research in “the right way” (see paragraph 2.1),
governments communicators need to understand – or at least have the support
of those who know and understand – the important principles, challenges and
processes involved.

2.7 SUMMARY

Since the researcher’s point of departure is that it is of critical importance that

communication research by government needs to be conducted on the basis of
sound theoretical principles and processes, a brief theoretic overview of research
in communication is provided in chapter 2. The concept communication research
is defined, and different types of research explained briefly. The research
process is attended to by means of a short discussion of the steps in the
research process. Lastly, the researcher describes some of the challenges of
communication research.

57