Chapter XIII: Descriptive Research Strategy
13.1 An Introduction To Correlational Research
• Five basic research strategies: experimental, non-experimental, quasi-experimental,
correlational, and descriptive.
• Descriptive research typically involves measuring a variable or set of variables as they
• The descriptive strategy is not concerned with relationships between variables but rather
with the description of individual variables.
• The goal is to describe a single variable or to obtain separate descriptions for each
variable when several are involved.
• Descriptive strategy is extremely useful in preliminary research.
• The intent of a descriptive strategy study is to describe a phenomenon.
• Three descriptive research designs are considered: observational research, survey
research, and case study research.
• In the observational research design, we describe observations of behaviors as they
occur in natural settings.
• In survey research design, we describe people’s responses to questions about behavior
• In case studies, we describe a single individual in great detail.
13.2 Observational Research Design
• In the observational research design, the researcher observes and systematically records
the behaviors of individuals for the purpose of describing behavior.
• As a measurement technique, behavioral observation can be used in a variety of
research strategies including experimental and correlational designs.
• A study using behavioral observations simply for descriptive purposes is classified as an
observational research design.
In the observational research design, the researcher observes and systematically records
the behavior of individuals in order to describe the behavior.
• The process of behavioral observation simply involves the direct observation and
systematic recording of behaviors, usually as the behaviors occur in a natural situation.
• This measurement technique, however, introduces two special measurement problems:
1. Because the goal is to observe natural behavior, it is essential that the behaviors
are not disrupted or influenced by the presence of an observer.
2. Observation and measurement require at least some degree of subjective
interpretation by the observer. The fact that the measurements are based, in
part, on a subjective judgment, raises the question of reliability; that is, would two
different occurrences of the same behavior be judged in the same way?
• The first problem can be addressed by concealing the observer so that the individuals do
not know their behaviors are being observed and recorded.
• As long as we observe public behaviors in public places, there is no ethical problems with
• An alternative procedure is to habituate the participants to the observer’s presence.
• Habituation requires repeated exposure until the observer’s presence is no longer a novel
• To address the second problem, subjectivity, researchers typically employ three
interrelated devices to help ensure the objectivity of their behavioral observations.
(i) a well-defined categories of behavior
(ii) inter-rater reliability (iii) use multiple observers to assess behaviors
• The first step in the process is to prepare a list of behaviors called behavior categories.
• Developing a set of behavior categories means that before observation begins, we
identify the categories of behavior we want to observe (such as group play, play alone,
aggression, social interaction) and then list exactly which behaviors count as examples of
• A pre-existing list enables observers to know exactly what to look for and how to
categorize each behavior.
• A set of pre-established behavior categories require a clear operational definition of each
construct being examined.
• During the observation period, normally only one individual observes and records
behaviors using the set of behavioral categories as a guide.
• To establish reliability, however, two or more individuals must observe and record
simultaneously for some of the observation periods.
• The degree of agreement between the two observers is then computed (usually as a
correlation or proportion of agreement ranging from 1.00 = perfect agreement to 0 = no
agreement) as a measure of inter-rater reliability.
• Behavioral observation involves converting the observations into numerical scores that
can be used to describe individuals and groups.
• The creation of numerical values is usually accomplished by one of three techniques:
(1) The frequency method involves counting the instances of each specific behavior that
occur during a fixed time observation period.
(2) The duration method involves recording how much time an individual spends engaged
in a specific behavior during a fixed-time observation period.
(3) The interval method involves dividing the observation period into a series of intervals
and then recording whether or not a specific behavior occurs during each interval.
• The first two techniques are often well-suited for specific behaviors but can lead to
distorted measurements in some situations.
• When an observer is confronted with a complex situation, it can be impossible to observe
many different individuals and record many different behaviors simultaneously.
• Solution #1: Use a videotape, which can be replayed repeatedly to gather observations.
• Solution #2: Take a sample of the potential observations rather than attempt to watch and
• The sampling process then consists of one of the following three procedures:
(1) Time sampling involves observing for one interval, then pushing during the next interval
to record all the observations. The sequence of observe-record-observe-record is
continued through the series of intervals.
(2) Event sampling involves identifying one specific event or behavior to be observed and
recorded during the first interval; then the observer shifts to a different event or behavior
during the second interval, and so on, for the full series of intervals.
(3) Individual sampling involves identifying one participant to be observed during the first
interval, then shifting attention to a different individual for the second interval, and so on.
Content Analysis & Archival Research
• The same techniques that are used in behavioral observation can be applied in other
situations that do not involve the direct observation of ongoing behaviors.
• It is possible to measure “behaviors” that unfold in movies or books, and it is possible to
study documents recording behaviors that occurred long ago.
• When researchers measure behaviors or events in books, movies, or other media, the
measurement process is called content analysis. • Recording behaviors from historical records is called archival research.
Content analysis involves using the techniques of behavioral observation to measure the
occurrence of specific events in literature, movies, television programs, or similar media that
present replicas of behaviors.
Archival research involves looking at historical records (archives) to measure behaviors or
events that occurred in the past.
• To ensure measurements are objective and reliable, the processes of content analysis
and archival research follow the same rules that are used for behavioral observation.
• The measurement process involves the following:
(1) Establishing behavioral categories to define exactly which events are included in each
category being measured. (i.e. a list of specific examples is prepared to define television
(2) Using the frequency method, the duration method, or the interval method to obtain a
numerical score for each behavioral category. (i.e. an observer records how many
examples of violence are seen in a 30 minute television program or how many
disciplinary actions appear on an individual’s school records)
(3) Using multiple observers for at least part of the measurement process to obtain a
measure of inter-rater reliability.
Types of Observation & Examples
• Ethologists are researchers who study nonhumans in their natural environment
• Researchers interested in human behavior commonly use the observational research
• There are three basic kinds of observation: naturalistic observation, participant
observation, and contrived observation.
• When a researcher observes and records behavior in a natural setting without intervening
in any way, it is called naturalistic observation or nonparticipant observation.
• A natural setting is one in which behavior ordinarily occurs and that has not been
arranged in any way for the purpose of recording behavior.
• In naturalistic observation, researchers try to be as inconspicuous and unobtrusive a
possible, passively recording whatever occurs.
In naturalistic observation or nonparticipant observation, a researcher observes behavior
in a natural setting as unobtrusively as possible.
• Naturalistic behaviors could be used to describe any behavior.
• Naturalistic observation is particularly useful in providing insight into real-world behavior.
• The results of studies using naturalistic observation also have high degrees of external
validity because the behavior is examined in real-world settings as opposed to
• Naturalistic observation is useful for examining behaviors that for practical or ethical
reasons cannot be manipulated by the researcher.
• One limitation of naturalistic observation is the time needed to conduct this type of
• A second problem with naturalistic observation is that the observer must take extra care
not to disrupt or influence the behavior because the goal is to observe natural behavior.
• In participant observation, a researcher does not observe from afar as one does in
• The researcher interacts with the participants and becomes one of them in order to
observe and record behavior. • This type of observation is needed in situations where inconspicuous observation is not
In participant observation, the researcher engages in the same activities as the people
being observed in order to observe and record their behavior.
• Participant observation allows researchers to observe behaviors that are not usually open
to scientific observation.
• By having the same experiences as the participants in the study, the observer gains a
unique perspective, obtaining insight into behavior not obtainable by observing from afar.
• The results of participant observation studies have high external validity because the
behaviors are examined in real-world settings, not laboratories.
• Limitation to this type of observation is that it is extremely time consuming.
• Participant observation is potentially dangerous for the observer.
• The observer may inadvertently alter participants’ behavior by directly interacting with
them; and, finally, by interacting with the participants and identifying closely with the
individuals in the study, and observer may lose objectivity.
• Contrived observation or structured observation – in contrast to observing behavior in
natural settings, the observer sets up a situation so that events can be observed without
waiting for them to occur naturally.
• The purpose of contrived observation is to precipitate a behavior that occurs naturally but
• The purpose of contrived observation is to create a situation wherein a natural behavior
may occur and be observed in a more timely fashion.
Observation of behavior in settings arranged specifically to facilitate the occurrence of
specific behaviors is known as contrived observation or structured observation.
• Often, such studies are conducted in laboratory settings.
• Developmental psychologists frequently use structured observation.
• Contrived observation may also take place in a natural but “set up” arena: a field setting
(which the participant perceives as a natural environment) arranged by the researcher for
the purposes of observing and recording a behavior.
• Structured observation is a compromise between the purely descriptive naturalistic
observation and manipulated field experiments.
• Ethologists frequently use contrived observation to study animals’ responses.
• Imprinting is the establishment of a strong, stable, preference for or attachment to an
object when that object is encountered during a sensitive period in an animal’s life.
• Konrad Lorenz discovered imprinting using naturalistic observation when the goslings
pursued him as if he were their parent. He and others then used contrived observation to
see if the young goslings would imprint on other models as well.
• Graylag goslings will imprint on almost any moving object in the environment.
• An advantage of contrived observation over both natural and participant observation is
that there is no waiting for behaviors to occur.
• The researcher makes them occur by setting up the environment in such a way that they
are more likely to occur.
• A disadvantage of contrived observation is that, because the environment is less natural,
the behavior may be as well.
Strengths & Weaknesses of Observational Research Designs
• A major strength of observational research is that the researcher observes and records
• In a survey research design, the researcher relies on the participants’ reports of their
behavior. • Participants can distort or conceal the accuracy or truthfulness of their responses, and
thus not reflect their actual behavior.
• Observational research results often have high external validity as well.
• With the exception of contrived observation in a laboratory, most observational research
is conducted in a field setting, and field research tends to have higher external validity.
• Another strength of observational research is its flexibility.
• A researcher can complete a comprehensive observation of antecedents, behaviors, and
consequences of the behaviors, whereas other studies examine a single, discrete
• A potential problem with observational research is the ethical concern about “spying” on
• If participants are not aware that their behavior is being observed, the researcher may be
violating a person’s privacy and right to choose to participate in the study.
• A weakness of all observational research designs is that they simply describe behavior
and do not examine its causes.
A Summary of the Strengths & Weaknesses of the Observational Research Design
Naturalistic Strengths Weaknesses
Observation • Behavior observed in the real • Time-consuming
world • Potential for observer
• Useful fore non-manipulated influence
behaviors • Potential for subjective
• Actual behaviors observed and interpretation
Participant Strengths Weaknesses
Observation • When natural observation is • Time consuming
impossible • Potential loss of
• Great information not objectivity
accessible otherwise • Increased chance for
• Participation gives unique observer influence
Contrived Strengths Weaknesses
Observation • Do not have to wait for • Less natural
behaviors to occur
13.3 Survey Research Design
• Surveys and questionnaires are used extensively in the behavioral sciences as relatively
efficient ways to gather large amounts of information.
• By presenting people with a few carefully constructed questions, it is possible to obtain
self-reported answers about attitudes, opinions, personal characteristics, and behaviors.
• The simple notion behind a survey is that it is not necessary to observe directly.
• With a survey, a researcher does not have to wait until a behavior or response occurs.
• Surveys can be used to obtain scores for a variety of different research designs.
• A survey often is conducted simply to obtain a description of a particular group of
• A study using a survey simply for destructive purposes is classified as a survey research
A research study that uses a survey to obtain a description of a particular group of individuals
is called a survey research design.
• The goal of the survey research design is to obtain an accurate picture of the individuals
• The survey provides a “snapshot” of the group at a particular time. • Sometimes, survey research focuses on specific characteristics.
• Other survey research may seek a more complex picture of a variety of behaviors and
• Depending on the questions asked, the results could provide a description on the item
• A common application of survey research is by companies to obtain more accurate
descriptions of their customers.
• Conducting survey research presents researchers with four issues that must be
addressed for the results to be accurate and meaningful.
• First, survey questions must be developed.
• Second, survey construction: the questions must be assembled and organized.
• Third, a selection process must be developed to determine exactly who will participate in
the survey and who will not.
• Survey participants must be representative of the general group to be studied.
• Fourth, administration of the survey.
The survey is used as a measurement technique in a variety of different research designs.
Simply because a study uses a survey does not mean that it is a survey research design. The
defining element of the survey research design is that the survey is used only to describe the
variables being studied.
Types of Questions
• There are different ways to ask participants for self-report information.
• Different types of questions encourage different types of responses.
• Different types of questions permit different types of responses.
• Different types of questions permit different degrees of freedom in the participants’
• The wording of a question also can introduce bias into participants’ answers.
• Different types of questions permit different types of statistical analysis and interpretation.
• If answers are limited to values on a nominal scale, you cannot compute a group
• An open-ended question simply introduces a topic and allows each participant to respond
in his own words.
• The primary advantage of an open-ended question is that it allows an individual the
greatest flexibility in choosing how to answer.
• An open-ended question imposes few restrictions on the participant and, therefore, is
likely to reveal each individual’s true thoughts or opinions.
• Although the question may lead the participant in a particular direction or suggest a
specific point of view, each individual is free to express her own thoughts.
• Disadvantage may be considered as different participants may approach the question
from entirely different perspectives, leaving the researcher with answers that are
impossible to compare or summarize.
• A second disadvantage of open-ended questions is that the answers are often difficult to
summarize or analyze with conventional statistical methods.
• Often, the researcher must impose some objective interpretation on the answers such as
classifying a rambling response as generally positive or generally negative.
• The responses to open-ended questions may be limited by a participant’s ability or
willingness to express thoughts.
• An inarticulate or tired person may give a very brief answer that does not completely
express the true breadth of her thinking. Restricted Questions
• A restricted question presents the participant with a limited number of response
alternatives, hence restricting the response possibilities.
• Like a multiple choice question, a restricted question typically asks the participant to
select the best or most appropriate answer in a series of choices.
• These questions produce a limited and predetermined set of responses.
• These questions are easy to analyze and summarize.
• Typically, the data are tabulated and reported as percentages or proportions of
participants selecting each alternative.
• It is also possible to obtain quantitative information from restricted questions by using an
ordered set of response alternatives.
• With restricted multiple choice questions, it is often possible to compute some kind of
average response for a group of participants.
• An element of open-endedness can be allowed in a restricted question by including a
blank category where participants are free to fill in their own responses (i.e. Other
(please specify ____________)
Rating Scale Questions
• A rating scale question requires a participant to respond by selecting a numerical value
on a predetermined scale.
• A scale from 1-10.
• The numerical scale that accompanies each question typically presents a range of
response alternatives from very positive to very negative.
• A common example uses a 5-point scale on which individuals rate their level of
agreement or disagreement with a simple statement [Strongly disagree, disagree, neither
agree or disagree, agree, strongly agree]
• The rating scale is usually presented as a horizontal line dividing into categories so that
participants can simply circle a number or mark X at the location corresponding to their
• This type of rating question is often called a Likert-scale or a Likert-type scale.
• Renis Likert developed the 5-point response scale as part of a much more sophisticated
• The scale is presented with equal spacing between the different response choices.
• The idea is to stimulate an interval scale of measurement.
• The responses from rating scales are usually treated as interval measurements.
• The distance between “agree” and “strongly agree” is treated as 1-point distance that is
equivalent to any other 1-point difference on the scale.
• There is no absolute rule for determining the number of categories for a rating scale
• Researchers commonly use from 5 to 10 numerical values.
• The re