Professional Documents
Culture Documents
interviews.
A. NATURE OF RESEARCH
• Ethnographic research is one form of
1. WHY RESEARCH IS OF VALUE
qualitative research. Other common forms of
• The scientific method provides an important qualitative research include the case study,
way to obtain accurate and reliable biography, phenomenology, and grounded
information. theory.
B. RESEARCH PROBLEM
C. LOCATING AND REVIEWING THE
• A research problem is the focus of a
RELATED LITERATURE
research investigation.
THE VALUE OF A LITERATURE REVIEW
RESEARCH QUESTIONS
• A literature review helps researchers learn
• Many research problems are stated as
what others have written about a topic. It
questions.
also lets researchers see the results of other,
• The essential characteristic of a related studies.
researchable question is that there be some
• A detailed literature review is often required
sort of information that can be collected in an
of master’s and doctoral students when they
attempt to answer the question.
design a thesis.
CHARACTERISTICS OF GOOD
TYPES OF SOURCES FOR A LITERATURE
RESEARCH QUESTIONS
REVIEW
• Research questions should be feasible—that
• Researchers need to be familiar with three
is, capable of being investigated with
basic types of sources (general references,
available resources.
primary sources, and secondary sources) in
• Research questions should be clear—that is, doing a literature review.
unambiguous.
• General reference tools are sources a
• Research questions should be significant— researcher consults to locate other sources.
that is, worthy of investigation.
• Primary sources are publications in which
• Research questions should be ethical—that researchers report the results of their
is, their investigation should not involve investigations. Most primary source material
physical or psychological harm or damage to is located in journal articles.
human beings or to the natural or social
• Secondary sources refer to publications in
environment of which they are a part.
which authors describe the work of others.
• Research questions often (although not
• The most common secondary sources in
always) suggest a relationship to be
education are textbooks.
investigated. The term relationship, as used
• Search terms are keywords or phrases • Researching the World Wide Web (WWW)
researchers use to help locate relevant should be considered, in addition to ERIC and
primary sources. PsycINFO, in doing a literature search.
STEPS INVOLVED IN A LITERATURE • Some of the information on the Web is
SEARCH classified into indexes, which group Web sites
together under similar categories. Yahoo! is
• The essential steps involved in a review of
an example of a directory.
the literature include:
• To obtain more specific information, search
(1) Defining the research problem as
engines should be used, because they search
precisely as possible;
all of the contents of a Web site.
(2) Deciding on the extent of the search;
THE LITERATURE REVIEW REPORT
(3) Deciding on the data base(s) to be
• The literature review report consists of an
searched;
introduction, the body of the review, a
(4) Formulating search terms; summary, the researcher’s conclusions, and a
bibliography.
(5) Searching general reference tools for
relevant primary sources; • A literature review should include a search
for relevant meta-analysis reports, as well as
(6) Obtaining and reading the primary
individual studies.
sources, and noting and summarizing key
points in the sources. • When a researcher does a meta-analysis,
he or she averages the results of a group of
WAYS TO DO A LITERATURE SEARCH
selected studies to get an overall index of
• Today, there are two ways to do a literature outcome or relationship.
search—manually, using print/paper tools to
locate print/paper sources; and electronically,
by means of a computer. The most common D. ETHICS AND RESEARCH
and frequently used method, however, is to
BASIC ETHICAL PRINCIPLES
search online, via computer. Regardless of
the tools involved, the search process is • Ethics refers to questions of right and
similar. wrong.
• There are five essential points (problem, • There are a number of ethical principles
hypotheses, procedures, findings, and that all researchers should be aware of and
conclusions) that researchers should record apply to their investigations.
when taking notes on a study.
• The basic ethical question for all
DOING A COMPUTER SEARCH researchers to consider is whether any
physical or psychological harm could come to
• Computer searches of the literature have a
anyone as a result of the research.
number of advantages—they are fast, are
fairly inexpensive, provide printouts, and • All subjects in a research study should be
enable researchers to search using more than assured that any data collected from or about
one descriptor at a time. them will be held in confidence.
• The steps in a traditional manual search are • The term deception, as used in research,
similar to those in a computer search, though refers to intentionally misinforming the
computer searches are usually the norm. subjects of a study as to some or all aspects
of the research topic.
• Plagiarism is the act of misrepresenting • A categorical variable varies only in kind,
someone else’s work as one’s own. not in degree or amount.
• Unintentional plagiarism can be avoided • Several kinds of variables are studied in
through the proper use and citation of educational research, the most common
published and unlisted sources. being independent and dependent variables.
RESEARCH WITH CHILDREN • An independent variable is presumed to
affect or influence other variables.
• Children as research subjects present
problems for researchers that are different • Independent variables are sometimes called
from those of adult subjects. Children are experimental variables or manipulated
more vulnerable, have fewer legal rights, and variables.
often do not understand the meaning of
• A dependent (or outcome) variable is
informed consent.
presumed to be affected by one or more
REGULATION OF RESEARCH independent variables.
• Before any research involving human • Independent variables may be either
beings can be conducted at an institution that manipulated or selected. A manipulated
receives federal funds, it must be reviewed variable is created by the researcher. A
by an institutional review board (IRB) at the selected variable is one that already exists
institution. that the researcher locates and then chooses
to study.
• The federal agency that has the major
responsibility for establishing the guidelines • A moderator variable is a secondary
for research studies that involve human independent variable that the researcher
subjects is the Department of Health and selects to study because he or she thinks it
Human Services. may affect the basic relationship between the
primary independent variable and the
dependent variable.
E. VARIABLES AND HYPOTHESES
• An extraneous variable is an independent
THE IMPORTANCE OF STUDYING variable that may have unintended effects on
RELATIONSHIPS a dependent variable in a particular study.
• Some of the more common threats to • The attitude of subjects toward a study
internal validity are differences in subject (and their participation in it) can create a
characteristics, mortality, location, threat to internal validity. This is known as
instrumentation, testing, history, maturation, subject attitude threat.
attitude of subjects, regression, and
• When subjects are given increased
implementation.
attention and recognition because they are
• The selection of people for a study may participating in a study, their responses may
result in the individuals or groups differing be affected. This is known as the Hawthorne
(i.e., the characteristics of the subjects may effect.
differ) from one another in unintended ways
• Whenever a group is selected because of
that are related to the variables to be
unusually high or low performance on a
studied.
pretest, it will, on average, score closer to the
• No matter how carefully the subjects of a mean on subsequent testing, regardless of
study (the sample) are selected, it is common what transpires in the meantime. This is
to lose some of them as the study called a regression threat.
progresses. This is known as mortality. Such
• Whenever an experimental group is treated
a loss of subjects may affect the outcomes of
in ways that are unintended and not a
a study.
necessary part of the method being studied,
• The particular locations in which data are an implementation threat can occur.
collected, or in which an intervention is
CONTROLLING THREATS TO INTERNAL
carried out, may create alternative
VALIDITY
explanations for any results that are
obtained. • Researchers can use a number of
techniques or procedures to control or
minimize threats to internal validity. • A standard score is a mathematically
Essentially they boil down to four derived score having comparable meaning on
alternatives: (1) standardizing the conditions different instruments.
under which the study occurs, (2) obtaining
TECHNIQUES FOR SUMMARIZING
and using more information on the subjects
QUANTITATIVE DATA
of the study, (3) obtaining and using more
information on the details of the study, and • A frequency distribution is a two-column
(4) choosing an appropriate design. listing, from high to low, of all the scores
along with their frequencies. In a grouped
frequency distribution, the scores have been
grouped into equal intervals.
J. DESCRIPTIVE STATISTICS • A frequency polygon is a graphic display of
a frequency distribution. It is a graphic way
STATISTICS VERSUS PARAMETERS
to summarize quantitative data for one
• A parameter is a characteristic of a variable.
population. It is a numerical or graphic way
• A graphic distribution of scores in which
to summarize data obtained from the
only a few individuals receive high scores is
population.
called a positively skewed polygon; one in
• A statistic, on the other hand, is a which only a few individuals receive low
characteristic of a sample. It is a numerical or scores is called a negatively skewed polygon.
graphic way to summarize data obtained
• A histogram is a bar graph used to display
from a sample.
quantitative data at the interval or ratio level
TYPES OF NUMERICAL DATA of measurement.