MOVING FROM A RESEARCH QUESTION

TO THE RESEARCH METHODOLOGY

Problem
After you have articulated your research question, you need to know the problem that is
behind the question. Chapter 4 of Booth et al. provides a very compelling case that it is
the problem that links your research to the audience. Your audience is the Research
Advisor, the Thesis Director, and the research community that is shared by his/her
interests and the problem that you characterize. Please understand this connection since
it needs to be reflected in the front section of your thesis proposal which is entitled,
Research Problem. This attention to the problem may lead you to restate the research
question.

The next part of the scientific method is to form a hypothesis. This is merely an educated
guess as to the answer for the question. You examine the literature on the subject;
scientists need libraries, reading is critical to scientific performance! You gather as much
book knowledge as you can on the subject to begin to arrive at an answer to your
question. This tentative answer...this best educated guess...is your hypothesis.

However, it is important to add a word about research approach first since the selection of
an approach will determine the role of the hypothesis. As you will find out, qualitative
and quantitative researchers tend to adopt different approaches to turn a topic to a focused
research question for a specific study. It is important to look at these differences before
getting into the hypothesis below.

Quantitative Research Qualitative Research

Test hypothesis that the researcher begins Capture and discover meaning once the
with researcher becomes immersed in the data
Concepts are in the form of distinct Concepts are in the form of themes, motifs,
variables generalizations, and taxonomies
Measures are systematically created before Measures are created in an ad hoc manner
data collection and are standardized and are often specific to the individual
setting or researcher
Data are in the form of numbers from Data are in the form of words and images
precise measurement from documents, observations, and
transcripts
Theory is largely causal and is deductive Theory can be causal or non-causal and is
often inductive
Procedures are standard and replication is Research procedures are particular and
assumed replication is very rare
Analysis proceeds by using statistics, Analysis proceeds by extracting themes or
tables, or charts and discussing how what generalizations from evidence and
they show relates to hypothesis organizing data to present a coherent,
consistent picture

1
Typical research questions for qualitative researchers include:

¾ How did a certain condition or situation originate?

¾ How is the condition/situation maintained over time?
¾ What are the processes by which a condition/situation changes, develops, or
operates?

The first two questions tries to confirm existing beliefs or assumptions. The last question
tries to discover new ideas.

Quantitative researchers take a linear path and emphasize objectivity. They are more
likely to use explicit, standardized procedures and a causal explanation. Their language
or variables and hypotheses is found across many areas of science that are based on a
positivist tradition. The process is often deductive with a sequence of discrete steps that
precede data collection: Narrow the topic to a more focused question, transform nebulous
theoretical concepts into more exact variables, and develop one or more hypotheses to
test. In actual practice, researchers move back and forth, but the general process flows in
a single, linear direction. In addition, quantitative researchers take special care to avoid
logical errors in hypothesis development and causal explanation.

Qualitative researchers follow a non-linear path and emphasize becoming intimate with
the details of a natural setting or a particular context. They use fewer standardized
procedures or explicit steps, and often devise on-the-spot techniques for one situation or
study. Their language of cases and contexts directs them to conduct detailed
investigations of particular cases or processes in their search for authenticity. They rarely
separate planning and design decisions into a distinct per-data collection stage, but
continue to develop the study design throughout early data collection. In fact, the more
inductive qualitative style encourages a slow, flexible evolution towards a specific focus
based on what the researcher is learning from the data. Grounded theory emerges from
the researcher’s continuous reflections on the data and the context.

Certainly the qualitative and quantitative distinction is overdrawn. But you will see that
many of the Harvard researchers potentially available as Thesis Directors will be
quantitative researchers. There is an opportunity to mix quantitative with qualitative
traits – this is called triangulation. It may be something you will want to consider. You
should also understand the research styles of the authors that will be prominent in your
literature research.

Arguments and Claims

In the Booth el al text, they refer to the hypothesis as a claim (Chapter 8). Chapter 7
covers the following topics:

Basing claims on reasons

Basing reasons on evidence

2
Acknowledging and responding to alternative claims
Warranting the relevance of reasons
Building complex arguments out of simple ones
Designing arguments not for yourself but for you readers: two common pitfalls.

Chapter 8 has a very easy to understand about claims (aka hypothesis) and qualifying
claims to enhance your credibility.

There are many ways to teach the scientific method. Some references do not have you
ask a question, but rather make a research statement. Although there are differences,
there are lessons to be learned from each approach to guide you when you set out to do
some original research. We are following the ALM Guide. So you need to see how
Booth et al complements and supports what is in the Guide even if the names are
different.

In the section below, I have garnered some information on hypotheses from other texts
and from articles published on the Internet. The hypothesis is perhaps the most important
part of preparing your thesis proposal. I want to make sure that the information is
complete and that we have an opportunity to discuss it in class and between you and
myself before you get too far with the other aspects of preparing yourself for the thesis
experience.

Hypothesis
Both a hypothesis and a problem contribute to the body of knowledge which supports or
refutes an existing theory. A hypothesis differs from a research question and problem.
A problem is formulated in the form of a question; it serves as the basis or origin from
which a hypothesis is derived. A hypothesis is a suggested solution to a problem. A
problem (question) cannot be directly tested, whereas a hypothesis can be tested and
verified.

A hypothesis is formulated after the problem has been stated and the literature study has
been concluded. It is formulated when the researcher is totally aware of the theoretical
and empirical background to the problem.

The hypothesis must be formulated in a way that provides the best answer to the research
question.

“What we have been considering is how to ask a question, have an idea, find a
topic suitable for research. It is a little difficult to separate that process from the
next step, development of a hypothesis, because the entire process—from first
idea to project definition, data gathering and analysis, right up to (and actually
after) the final punctuation mark on the final page of the thesis—is continuous.
But let us restrict ourselves in this section to actual development and formulation
of the hypothesis.” (ALM Guide, page 16)

3
A hypothesis grows out of the research question. It is an assertion, conjecture or
premise, subject to verification via research. It is the consequence of organizing our
questions and other information and the expected answers that grow out of them.

Moriarty (page 10) states,

“A hypothesis does not emerge from out of thin air. It is formulated as a possible
answer to the question. A hypothesis can be an educated guess about the possible
answers to the question or a supposition about anticipated results. In order to
devise a test – to select measuring devices, substances, organisms, time frame,
computer access – a reasonable sense of the way results should/might/could turn
out is framed. To formulate an educated guess or a supposition, one must have
some familiarity about the subject to inform the question that is posed.”

Please notice that hypotheses do not always have to be correct. In fact most of science is
spent trying to determine the validity of a hypothesis, yet this effort is NOT likely to give
a single perfect answer. So, in formulating your hypothesis, you should not worry too
much that you have come up with the best or the only possible hypothesis. The rest of the
scientific method will test your hypothesis. What will be important is your decision at the
end of the method.

The purpose of a hypothesis:

• It offers explanations for the relationships between those variables that can be
empirically tested.
• It furnishes proof that the researcher has sufficient background knowledge to
enable him/her to make suggestions in order to extend existing knowledge.
• It gives direction to an investigation.
• It structures the next phase in the investigation and therefore furnishes continuity
to the examination of the problem.

The characteristics of a hypothesis:

• It should have elucidating power.

• It should strive to furnish an acceptable explanation of the phenomenon.
• It must be verifiable.
• It must be formulated in simple, understandable terms.
• It should correspond with existing knowledge.

Research hypotheses are the specific testable predictions made about the independent and
dependent variables in the study. Usually the literature review has given background
material that justifies the particular hypotheses that are to be tested. Hypotheses are
couched in terms of the particular independent and dependent variables that are going to
be used in the study.

4
From the literature it is concluded that a hypothesis is a tentative statement, that implies a
proposed answer to a research question and problem, setting accountability and
responsibility of effective research procedure as high priority..

It should, however, be emphasized that in no way a hypothesis statement can claim to be

the only solution to the research question and problem. It serves only as a point of
departure - the chain between the theory and the research that leads to the broadening of
knowledge.

Hypotheses are thus tentative statements that should either be acknowledged or rejected
by means of research. It really must be rejectable. There must be a way to test the
possible answer to try to make it fail. If you design an untestable hypothesis, then science
cannot be used to help you decide if it is right or not. For the moment, let us say that your
question is "Is God awake?" and you have made the hypothesis "God is awake." There is
no way to test the slumbering state of God scientifically.

Because hypotheses give structure and direction to research, the following aspects should
be kept in mind when formulating a hypothesis:

• Hypotheses can only be formulated after the researcher has gained enough
knowledge regarding the nature, extent and intensity of the problem.
• Hypotheses should figure throughout the research process in order to give
structure to the research.
• Hypotheses are tentative statements/solutions or explanations of the formulated
problem. Care should be taken not to over-simplify and generalize the formulation
of hypotheses.
• The research question and problem do not have to consist of one hypothesis only.
The type of problem area investigated, the extent which encircles the research
field are the determining factors on how many hypotheses will be included in the
research proposal.

A research hypothesis is usually stated in an explanatory form, because it indicates the

expected reference of the difference between two variables. ln other words it verifies the
reference that the researcher expects by means of incorporating selected research
procedures.

The research hypothesis may be stated in a directional or non-directional form. A

directional hypothesis statement indicates the expected direction of results, while a non-
directional one indicates no difference or no relationship.

It is a relationship between variables and indicates the nature of the relationship.

If A is valid, B follows ...
If you hit a child with a stick, he/she will cry.
Schools in which pupil-teacher relations are open/friendly will have less unrest than
comparable schools where pupil-teacher relations are closed/tense.

5
The hypothesis converts the research question posed into a declarative statement that
predicts an expected outcome. It identifies a predicted relationship between two or more
variables.

Null Hypothesis states that there is no relationship between the variables

Research consists of a statement about the expected relationship

Hypothesis between the variables
Directional Hypothesis direction of the relationship is specified: greater
than, less than, positively, negatively

Non-directional a relationship is specified, but the direction is not

Hypothesis specified

The statement that follows has been written as a research hypothesis and a null
hypothesis:

Aggressiveness of nursing care and DNR orders:

Research hypothesis: Aggressive nursing care is reduced when caring for a

patient with a DNR order.

Null hypothesis: There is no difference in the aggressiveness of nursing

care when caring for a patient with a DNR order as
compared to a patient with no DNR order.

ln order to assist you in formulating a research hypothesis, you should consider the
criteria that should be used in formulating hypotheses:

The following criteria are of importance in formulating hypotheses. A hypothesis should:

• Stand a test;
• Be expressed in clear language;
• Be in accordance with the general theme of other hypotheses statements in the
same field of study, and should be regarded as valid;
• Be. coordinated with the theory of science;
• Be a tentative answer to the formulated research question and problem;
• Be logical and simplistic;
• Consider available research techniques (to be able to analyze and interpret the
results);
• Be specific; and
• Be relevant to the collection of empirical phenomenon and not merely conclude
value judgments.

6
Evaluating the Hypothesis
What are the criteria for evaluating hypotheses? We can use three criteria:
correspondence, coherence, and conceptual elegance. By correspondence we mean
correspondence to the available, relevant source testimony; by coherence we mean a
logical, well-focused, internally consistent hypothesis; and by conceptual elegance we
mean the absolute minimum of abstract constructs and unstated assumptions within the
hypothesis.

In several ways, formulation of the hypothesis and research protocol or methods is the
most difficult part of the entire thesis process. Since it serves as the outline, flow chart,
or general recipe for the whole project, it can scarcely be too thorough. This does not
mean that it must be immense. Rather, it must be clear, precise, and succinct; it must be
defensible on theoretical grounds and logistically feasible; and it must honestly consider
its own shortcomings and limitations. That means substantial research must be completed
before it will be possible to construct even an initial hypothesis

“A research problem involves not simply a tightly focused question and original research,
but an answer to that question based on original thinking. You should try to make a well-
reasoned point to substantiate the hypothesis advanced in your thesis.” (ALM Guide,
page 41)

It can be equally productive to look for evidence that opposes as well as supports your
hypothesis and to deal with this information directly—e.g., research studies that seem to
contradict your own data. If the hypothesis remains credible in the face of opposing
evidence, you can defend your argument with perhaps even greater confidence and
authority. Think of yourself not so much as a lawyer arguing a case as a judge evaluating
the evidence on both sides of an issue in a trial. Through this kind of disinterested
exploration of a question, the skilled researcher arrives at a position of relative certitude
and is thus more likely to persuade the reader of the validity of his or her conclusions.

Moriarty (page 13) states,

“To test a hypothesis means to subject it to trial. Positive evidence produced by

testing confirms the hypothesis; negative results produced by the test disconfirm
the hypothesis. A hypothesis functions in two domains: (1) it is a statement made
in language and (2) it is an assertion about the world. The domain delimits the
kind of evidence used to confirm or disconfirm the hypothesis. (1) The
hypothesis, considered as a speculation framed in language, can be validated by
other language statements…. (2) As a speculation about events in the world, a
hypothesis can be validated by empirical evidence. However a hypothesis must
be phrased in such a way as to render it testable…Some hypotheses can be
validated by both empirical evidence and language statements.”

Moriarty goes on to say (page 14):

7
 “The object of experimentation is to falsify the hypothesis. If you test a
hypothesis with the intention of proving that it is false, you will apply more sever
and rigorous tests. If a hypothesis passes many severe tests that attempt to show
that it is false, that hypothesis may be said to be corroborated or verified.”

A hypothesis may be assessed at four different levels (cited by Moriarty, pages 14-15):

1. “Logical consistency. Is the hypothesis logically consistent and not sefl-

contradictory? Is it framed in the context of a logical approach? Because
hypotheses grow out of a received body of knowledge and expectations about
what should happen, does this hypothesis reflect the most adequate statement or
response to the research question?
2. Explanatory value. Does the hypothesis promise to offer some explanatory value
by making the observed phenomena intelligible? In other words, does the
hypothesis promise to clarify, explain, or suggest an interpretation? What can the
hypothesis explain? What advantage does this hypothesis offer over other
possible theories? What are the reasons for discarding alternative hypotheses?
3. Relation to other theories. How does the hypothesis compare with other
hypotheses or theories current in the field? Does it merit testing? Has it been
tested before? Does it represent an advance or a new insight with respect to
alternative theories? (Inconsistency with other theories is no reason to reject it.)
Given two hypothesis of equal explanatory value, which is more simple – i.e.,
which contains the fewer irreducible assumptions?
4. Empirical demonstration. Does empirical evidence bear out the hypothesis? Can
empirical evidence bear it out or must validation be found another way? Is the
hypothesis capable of being tested (falsified)?”

An experiment is expected to reveal causal relations. To assess causal relations,

controlled conditions are stipulated to limit the number and function of independent
variables in order to focus on the ones under investigation.

Causal Theory and Hypotheses

[Much of this section is taken from Neuman (2003)]
A hypothesis is a proposition to be tested or a tentative statement of a relationship
between two variables. Hypotheses are guesses about how the world works, they are
stated in a neutral form.

“Hypotheses are much more important in scientific research than they would
appear to be jus by knowing what they are and how they are constructed. They
have a deep and highly significant purpose of taking man our of
himself…Hypotheses are powerful tools for the advancement of knowledge,
because, although formulated by people, they can be tested and shown to be
correct or incorrect apart from a person’s values and beliefs.”

8
Here are five characteristics of causal hypotheses:

1. It has at least two variables.

2. It expresses a causal or cause-effect relationship between the variables.
3. It can be expressed as a prediction or an expected future outcome.
4. It is logically linked to a research question and a theory.
5. It is falsifiable; that is, it is capable of being tested against empirical evidence and
shown to be true or false.

The first two characteristics define the minimum elements of a hypothesis. The third
restates the hypothesis. For example, the hypothesis that attending religious services
reduces the probability of divorce can be restated as a prediction: “Couples who attend
religious services frequently have a lower divorce rate than do couples who rarely attend
religious services. The prediction can be tested against empirical evidence. The fourth
characteristic states that the hypothesis should be logically tied to a research question
and to a theory. Researchers test hypotheses to answer the research question or to find
empirical support for a theory. The last characteristic requires that a researcher use
empirical data to test the hypothesis. Statements that are necessarily true as a result of
logic, or questions that are impossible to answer through empirical observation cannot be
scientific hypotheses!

Causal hypotheses can be stated in several ways. Sometimes the word cause is used, but
this is not necessary. Here are some ways to state causal relations:

¾ Religious attendance causes reduced divorce.

¾ Religious attendance leads to reduced divorce.
¾ Religious attendance is related to reduced divorce.
¾ Religious attendance influences the reduction of divorce.
¾ Religious attendance is associated with reduced divorce.
¾ Religious attendance produces reduced divorce.
¾ Religious attendance results in reduced divorce.
¾ If people attend religious services, then the likelihood of divorce will be reduced.
¾ The higher religious attendance, the lower the likelihood of divorce.
¾ Religious attendance reduces the likelihood of divorce.

Researchers avoid using the term proved when testing hypotheses. You might hear the
term proof used in journalism, courts of law, or advertisements, but you will rarely hear
research scientists using it. Scientists do not say they have proved a hypothesis or a
causal relationship. Proof implies finality, absolute certainty, or something that does not
need further investigation. Proof is too strong a term for that cautious world of science.
Evidence supports or confirms, but does not prove, the hypothesis. Scientists do not want
to close off the possibility of discovering new evidence that might contradict past
findings. They do not want to cut off future inquiry or stop exploring intervening
mechanisms. Proof is used when referring to logical or mathematical relations, as in a
mathematical proof, but not in discussing empirical research.

9
Hypotheses are links in a theoretical causal chain and can take several forms.
Researchers use them to test the direction and strength of a relationship between
variables. When a hypothesis defeats its competitors, or offers alternative explanations
for a causal relation, it indirectly lends support to the researcher’s explanation. A curious
aspect of hypothesis testing is that researchers treat evidence that supports a hypothesis
differently from evidence that opposes it. They give negative evidence more importance.
The idea that negative evidence is critical when evaluating a hypothesis comes from the
logic of disconfirming hypotheses (related to the idea of falsification and with the use of
the null hypothesis).

Negative and disconfirming evidence shows that the predictions are wrong. Positive or
confirming evidence for a hypothesis is less critical because alternative hypotheses may
make the same prediction. A researcher who finds confirming evidence for a prediction
may not elevate one explanation over its alternatives.

Researchers test hypotheses in two ways: a straightforward way and a null hypothesis
way. Many quantitative researchers, especially experimenters, frame hypotheses in terms
of a null hypothesis based on the logic of the disconfirming hypotheses. They test
hypotheses by looking for evidence that will allow them to accept or reject the null
hypothesis. Most people talk about a hypothesis as a way to predict a relationship. The
null hypothesis does the opposite. It predicts NO relationship. Researchers use the null
hypothesis with a corresponding alternative hypothesis or experimental hypothesis. For
example, I believe that students who live on campus in dormitories get higher grades than
students who live off campus and commute to college. My null hypothesis is that there is
no relationship between residence and grades. The alternative hypothesis is that students’
on-campus residence has a positive effect on grades.

For most people, the null hypothesis approach is a backward way of hypothesis testing.
Null hypothesis thinking rests on the assumption that researchers try to discover a
relationship, so hypothesis testing should be designed to make finding a relationship
more demanding. A researcher who uses the null hypothesis approach only directly tests
the null hypothesis. If evidence supports or leads the researcher to accept the null
hypothesis, he or she concludes that the tested relationship does not exist. This implies
that the alternative hypothesis is false. On the other hand, if the researcher can find
evidence to reject the null hypothesis, then the alternative hypotheses remain a
possibility. The researcher cannot prove the alternative: rather, by testing the null
hypotheses, he or she keeps the alternative hypotheses in contention. When null
hypothesis testing is added to confirming evidence, the argument for an alternative
hypothesis can grow stronger over time.

A common approach to the hypothesis in environmental management is the

argumentative approach. Argumentative papers take a persuasive approach. The
hypothesis is developed, and then data and information are gathered to buttress or support
the thesis. Argumentative papers use the deductive method, that is, the hypothesis directs
the data collection or literature review. Data are analyzed in the context of the point to be
made. The deductive method:

10
 o Moves from the general statement of belief to specific evidence supporting
that belief(s)
1. Hypothesis
2. Data collection to support the hypothesis
3. Analysis of data with interpretation intended to support you position
4. Arguments tied to evidences to support the thesis
5. Convincing conclusion using evidence to support a stance on a
debatable issue
o Searches (and researches) for evidence only to logically buttress the
argument
o Sustains the argument throughout the paper
o Convinces and reaches definite closure on the issue

The argument is the thesis.

To recapitulate, an ALM thesis asks a question and answers it with a proposition or

hypothesis supported by evidence, which may or may not include empirical research, but
must include original thought. What do we mean by “original thought”? We do not
mean that you must think of something that no one else has ever thought of before. We
do mean that you should come up with a hypothesis and offer supporting evidence in
the thesis that does not rely solely on the authority of others to support your point. The
argument, in other words, involves an interpretation that is your explanation of why
things are the way they are and offers evidence that you have gathered, the sum total of
which makes a contribution to our knowledge and understanding of a particular subject.

In this context, ALM candidates should understand the research basis of the ALM
degree. The ALM degree is a liberal arts degree, not a professional degree that might
allow more latitude for creative or applied components. The ALM thesis therefore must
involve a research problem.

Now, during the formal collecting of data, as during the preliminary questions and
development of the hypothesis, matters rarely proceed so smoothly as you might like. At
any step, it is possible and actually likely that data or answers will differ from those we
expected. Many things can happen during research. Perhaps we discover that someone
else has already tested our hypothesis or answered our questions. Perhaps the hypothesis
was based on a misunderstanding or an erroneous assumption, either our own or one
commonly held by researchers in that discipline. Perhaps we discover something
unexpected, which completely changes the research situation.

General Approach to Academic Inquiry

For the following discussion, we will divide research into two basic types: bibliographic
work and empirical research, either observational or experimental. This distinction is
somewhat artificial, because we are limiting the former to background reading and
collecting an overview of the chosen field. Certainly empirical bibliographic research can

11
be carried out. Every ALM thesis will include at least the first; virtually all will include
some aspects of the second. We will discuss each in turn and some logistical matters as
well.

Bibliographic Research
“Such work is here considered to be the background preparation for formulation and
testing of a hypothesis, involving printed matter, whether consisting of primary or other
sources. In other words, this is the literature-based reading and thinking essential to the
development and elaboration of a question and hypothesis. It can be a way of finding
your question, as well as developing the next stages once you already have a question.”
(ALM Guide, page 18)

All thorough investigation includes a careful analysis and critique of the published
sources related to your question. The Proseminar Instructor or research advisor may be
especially helpful here, but one of the student’s chief tasks is to acquire a thorough
knowledge of the literature relevant to the research problem. This knowledge is
mandatory in all areas of research. Any responsible exploration of a research topic
requires an understanding of others’ investigations as they appear in the relevant
literature. Your job is, first, to master existing knowledge in your field and then to extend
that knowledge with an original contribution.

It should be emphasized again that a simple overview or description of the literature is

not sufficient or appropriate as a research project. If, for instance, all the literature on a
subject is based on a false premise, and the investigator can demonstrate this, then a
critical review is indeed an appropriate research problem. But such a discovery is most
uncommon. Usually, a review of the literature is only the first stage in an investigation. It
is the preamble to a hypothesis and an analysis or interpretation designed to add
something to the special field of knowledge. You will first refer to secondary sources:
books written about a particular subject that are a distillation of the author’s research.
However, is expected that you will be reading almost entirely primary sources. By
definition, a secondary source is based on someone else’s opinion. To some extent, you
must trust the author; but the careful researcher will do well to compare one secondary
source with another, examining the source material of each while looking for consensus.
You should not necessarily trust blindly the first “expert” whose work you happen to read
or even the majority of experts.

In using any literature, the researcher is reminded to pay particular attention to how
recently it has been published and how valid and reliable it is, and in some cases to the
scholarly reputation of the author or publisher. ALM candidates are expected to be
familiar with the most recent publications in their field. And, while they may not be
sufficiently expert to judge the reliability of a piece of research, they must read enough so
that they are sufficiently familiar with the standard works in the field to recognize an
unconventional analysis and obtain assistance in evaluating it.

In developing a complete list of relevant materials, candidates should perform literature

searches. Depending upon the topic, these may be done using electronic databases, print

12
resources, or both. Research librarians can help candidates decide whether and when to
undertake a search, as well as which databases and print resources would be best to use.
Of course, a well-constructed search will quickly let you know about the novelty of your
topic, as well as the amount of research material available.

Information about relevant databases, both online and on CD-ROM can be obtained from
Dr. George Clark. Most of the electronic databases are available through the Harvard
University Libraries website. If you have not yet been accepted into the ALM program,
and thus do not have a University ID or PIN, access to the indexes, abstracts, and so on,
is available from the Grossman Library.

Students should familiarize themselves with the University Libraries system as soon as
possible, as it is large and complicated, with over 90 libraries containing over 14 million
books, journals, and other materials. Depending upon your area of research, you could
find that most of the material you need is in one library; you could also find that you need
to do research in four or five. The major reference library for environmental sciences is
the Cabot Science Library. The largest of Harvard’s libraries is Widener, the main library
for the study of the humanities and social sciences, which itself contains over 3 million
books, journals, and other materials. Widener’s Research and Bibliographic Services
Department offers tours, which are very useful. Even if you have completed a proseminar
and had experience using Widener already, a “refresher” tour may be helpful as you
begin your research. It is also extremely important to learn how to use HOLLIS, the
University Libraries’ catalog, as well as the journal indexing and abstracting resources,
the various encyclopedias, dictionaries, and bibliographies in your field of interest, such
as the MLA International Bibliography, the Humanities and Social Sciences Index, and
the Reader’s Guide to Periodical Literature. Students should also consult Dissertation
Abstracts, available online through the University Libraries site, to discover whether
unpublished theses on their proposed topic have already been completed.

Costs for computer searches vary with the database, field of specialization, and access
location. In some cases there may be more than one way to gain access to a database, and
the cost may differ dramatically depending on how the information is accessed. More
information about costs and specific procedures can be obtained from the reference
librarian. Many information sources on CD-ROM are available free of charge in the
reference room of Widener, where a brochure can be obtained that introduces researchers
to some of these. Additional on-line information databases now exist in virtually all
fields, some of which are free and can be accessed from Grossman Library, while others
require subscription. Harvard Extension School does not supply funds for this purpose.

Empirical Research
Empirical research serves to extend, challenge, or verify a hypothesis by obtaining
original data through direct observation and/or experimentation. In the environmental
sciences, direct observation means that the researcher observes behavior or phenomena
(associated with chemicals, humans, technology, whatever) directly. Its parallel in the
humanities and social sciences is working with primary sources—living subjects and

13
original documents. These can even be mixed in some cases. (ALM Guide, page 19 –
21)

Direct observation can be naturalistic, experimental, or quasi-experimental. The first

means that the investigator approaches an event as an observer and makes no deliberate
efforts to alter the situation. It occurs in daily experience, making no intrusions upon it.
The second and third involve experimental manipulation, with the distinction between
experimental and quasi-experimental designs involving the random assignment of cases
to conditions in true experiments and the use of naturally occurring groups in the latter.
Many behavioral science investigations require a quasi-experimental design because
some characteristics of interest cannot be randomized. For example, subjects come to an
investigation already male or female, or older or younger.

In making naturalistic observations, the goal is to observe as closely and meticulously as

possible, using precise, carefully operationalized definitions and specified observation
and measurement techniques. In both the sciences and the humanities, the issues of scope
of the observations, threats to validity such as investigator bias, and reactivity must be
anticipated and dealt with at the outset in order to be sure that results will be both reliable
and valid. Constructing an experiment or making observations that will yield credible and
useful results is a complicated business. The researcher must take special care not to
accidentally subvert the conclusiveness and objectivity of the results. Let us consider
each of these in detail:

(1) “Scope of the observations” includes the breadth of the question: all the pieces of
sculpture done by one woman artist, a comparison of the works of several women artists
from New York City over a fifty-year period, or selected works that depict similar subject
matter by only one artist? This is not just a question of how much material will be
covered; it includes also the direction of the research. Each of the three options presented
has a different slant. And for each slant, there is a necessary minimum (as well as a
feasible maximum) amount of material to be considered. In planning the thesis, students
should be aware of pragmatic issues that may interfere with obtaining sufficient data.
Cost or time is not a sufficient reason to draw conclusions based on inadequate samples
or inappropriate methods.

In the environmental, biological, social, and behavioral sciences, this is directly related to
the sample-size question. It boils down to whether the researcher has an adequate and a
representative amount of data for the number of variables to be considered. The question
to be answered and the type and amount of data collected must match. Detailed, extensive
observations of five subjects allow for very different conclusions than more specific and
precise measurements of one or two dimensions on 100 subjects. Five subjects are
sufficient for a case series using applied behavior analytic methods. One hundred subjects
allow for an examination of group differences or co-variations among several specific
variables, e.g., an examination of how genotype predicts phenotypic expressions of a
particular characteristic. Trade-offs exist between intensive case studies of just a few
individuals and less intensive group studies of a larger population. Each method is
appropriate to different kinds of questions and the quality and depth of observations

14
required for analysis. One of the things you will learn as you familiarize yourself with
the literature in any field is what constitutes a typical sample size.

(2) “Investigator bias” refers to any effect on the research due to the experiences,
prejudices, interests, and so forth of the researcher or investigator. Since it is effectively
impossible for us to be free of bias, the best we can do is be aware of our biases and
attempt to control or compensate for them. A researcher with a vested interest in some
particular outcome to the research may be markedly handicapped in interpreting the
evidence. Someone violently opposed to the Castro regime in Cuba, for example, may
find it impossible to be sufficiently objective to do thorough research on that country.
Particularly but not only in experiments, the researcher has expectations of results based
upon the hypotheses and experimental manipulations. To control this, experiments are
typically run “blind,” that is, without the investigator knowing which subjects were
assigned to which experimental condition until all data have been coded.

On another and more common level, the researcher may be working with documents that
are themselves biased. US Civil War diaries will show very different reactions to and
interpretations of given events, depending on whether the author is from the North or
South. Note, however, that this does not diminish the value of such diaries. On the
contrary, it is most useful to have more than one outlook or perspective on any matter or
event. But the researcher must be aware of these potential biases and be able to deal with
research materials accordingly.

(3) “Investigator effects” differ from investigator bias in that the latter refers to
influences due to the investigator’s mindset, while the former refers to influences due to
unintended effects of procedures, to the observer’s physical presence, and to
manifestations of the observer. These effects are due to subject reactivity and are
particularly relevant when working with human subjects or animals. In essence, we all
behave differently when we know we are being observed, and in designing an
investigation it is the researcher’s responsibility to mitigate this effect to the extent
possible. This may mean giving subjects time to acclimate to the experimental situation
or observing from some distance (e.g., using a one-way mirror or videotaping a subject
from outside the room). In studies of hypertension, “white coat hypertension” refers to an
elevation in blood pressure that occurs in many people when their blood pressure is
assessed in a medical setting. To control for this effect, it is expected that a subject will
be asked to rest for 5 minutes prior to the assessment and then the measurement will be
taken 3 times, allowing the subject to get used to the experimenter, the equipment, and
the sensations engendered by the procedure. Again, it is virtually impossible to eliminate
such investigator effects, but a responsible researcher will eliminate as much as possible,
and keep the remainder constant. Make conditions minimally disruptive, keep yourself as
low-key and neutral as possible, be alert to uneasiness displayed by the subject.

For human subjects, this might mean giving them time to get accustomed to the
researcher, speaking their native language, dressing neutrally, making observations from
behind a one-way mirror. For other subjects, orientation or “warm-up” time (that is,
giving them time to get used to the researcher) is primary. Some animals, like ants,
require no time to acclimate to the experimental or observational environment; while

15
others, like primates, can take months or even years. Make observation conditions as
unobtrusive as possible—after all, you want to maximize the amount of normal behavior
you will see. Depending on the animal, use as little light as possible, dress in a typical,
recognizable way (though you need not dress specially for, say, ants!), be calm and quiet.

Methodology and Methods

The approach a researcher uses to investigate a subject is termed the methodology.
Methodology refers to the philosophical basis on which the research is founded. The
particular techniques used to collect data and information are termed methods. Don’t
become confused by these two terms – they are NOT the same.

Research is a process, a series of activities unfolding over time. It involves a number of

things such as choice of the research design, the collection of data, the evaluation of
results and the formulation of conclusions and more questions.

A criticism of made by many thesis directors is that the research methodology lacks
clarity and direction. In many instances the method of data collection is not well thought
out or suitable to the subject being investigated. It may look like the student has become
impatient and simply wants to chase results and information. You must remember, how
you collect evidence is as important as the evidence itself. If your results are to be
believed, then the way you collect them in the first place must also be believable.

Methodology is concerned with how the researcher views the world in which he or she
carries out the research. There are two principle methodologies available to you:
quantitative research and qualitative research. Quantitative research involves an
objective way of studying things whereas a qualitative approach assumes that this is
difficult and the research is subjective.

Quantitative research is sometimes referred to as positivist. It aims to be objective and

collects and uses numerical data. This data may be collected by the researcher or may
already exist and is used for another purpose. The researcher uses a mathematical and
statistical treatment to help evaluate the results. Environmental scientists carrying out
experiments use this approach. Surveys using questionnaires and interviews where
responses are given numerical values would also be described as quantitative research.

Qualitative research takes the view that it is very difficult for research to stand back and
be objective, since they are really part of the problem being researched. This type of
research is also called relativist or phenomenalist. With this approach, data is usually
collected in the form of descriptions. Even though some of the methods used, such as
interviews, are used in quantitative research, the difference is that qualitative researchers
only use non-mathematical procedures when interpreting and explaining their research.

It is possible to use BOTH approaches in research. You might decide to carry out a
survey after investigating a subject from a qualitative perspective. For example, after
studying certain changes in a subject area, you may decide to survey, by questionnaire,

16
opinions about a particular aspect associated with the changes. Looking at the same
problem from a number of viewpoints is an excellent way to verify your interpretation
and conclusions. Using a number of different approaches is termed triangulation.

Academics argue about the merits and limitations of each approach and sometimes even
over what each term really means. You should adopt the methodology that best suits
your background, interest, and, most importantly, the subject you are investigating. I will
not address the various methods of data collection associated with these methodologies at
this point.

Research Design
Research design is a general term that covers a number of separate, but related, issues
associated with your research. It includes the aims of the research, the final selection of
the appropriate methodology, the data collection techniques you intend to use, the chosen
methods of data analysis, and interpretation. The design is how all of this fits together.

Two important concepts to consider in your research design are validity and reliability.
Validity is concerned with the idea that research design fully addresses the research
hypothesis you are trying to prove to be correct. This implies that as much planning as
possible must be done beforehand. Reliability is about consistency and research, and
whether another researcher could use your design and obtain similar findings. This does
not imply that their interpretation and conclusions would be the same. The chances are
that they will be different since this is where the judgment of individual researches comes
into play.

Overall, the research design is the blueprint or detailed outline for the whole of your
research and thesis. It relies on careful prospective planning.

There will be more specific information on methodology and methods in the next few
weeks.

17