Psychology Textbook .pdf
Original filename: Psychology Textbook.pdf
This PDF 1.6 document has been generated by Adobe Acrobat Pro 11.0.6 / Adobe Acrobat Pro 11.0.6 Paper Capture Plug-in, and has been sent on pdf-archive.com on 21/11/2014 at 06:14, from IP address 76.228.x.x.
The current document download page has been viewed 2691 times.
File size: 12.6 MB (369 pages).
Privacy: public file
Download original PDF file
SL and HL
• Evaluate research studies.
Theory and practice in qualitative research
• Distinguish between qualitative and quantitative data.
• Explain strengths and limitations of a qualitative approach to research.
• To what extent can findings be generalized from qualitative studies?
• Discuss ethical considerations in qualitative research.
• Discuss sampling techniques appropriate to qualitative research (for example,
purposive sampling, snowball sampling).
• Explain effects of participant expectations and researcher bias in qualitative research.
• Explain the importance of credibility in qualitative research.
• Explain the effect of triangulation on the credibility / trustworthiness of qualitative
• Explain reflexivity in qualitative research.
• Evaluate semi-structured, focus group and narrative interviews.
• Discuss considerations involved before, during and after an interview (for example,
sampling methods, data recording, traditional versus postmodern transcription,
• Explain how researchers use inductive content analysis (thematic analysis) on
• Evaluate participant, non-participant, naturalistic, overt and covert observations.
• Discuss considerations involved in setting up and carrying out an observation (for
example, audience effect, Hawthorne effect, disclosure).
• Discuss how researchers analyse data obtained in observational research.
• Evaluate the use of case studies in research.
• Explain how a case study could be used to investigate a problem in an organization
or group (for example, a football team, a school, a family).
• Discuss the extent to which findings can be generalized from a single case study.
HL vs SL: What you need to know
The IB Psychology syllabus expects all students to develop a good understanding of the
most common methods used in psychological research. There are two ways to achieve this:
• requiring students to describe, explain and evaluate published research
• requiring students to carry out their own research for the Internal Assessment.
In addition to this, Higher Level students are also required to complete a separate
examination paper relating to qualitative methods.
What is good research? Reliability and
Researchers in psychology have traditionally attempted to adopt a scientific approach,
whether trying to add to understanding of mind and behaviour through the creation and
testing of theories, or trying to answer a practical question like how to increase motivation
at work or in sport. Efforts to follow a scientific approach have helped the academic
community to accept findings made by psychological researchers as a valid form of
knowledge, or in other words, as ‘true’.
Empiricism and objectivity
Empirical evidence is
evidence gathered using
our own senses.
For psychological research to be considered scientific, several key ideas need to be
embraced. Perhaps the most fundamental of these is empiricism: an approach to the
acquisition of knowledge that places high value on direct sensory information. It is not
enough for psychologists to make knowledge claims based solely on their own thoughts or
beliefs; there must be empirical evidence for their ideas. The evidence is gathered using the
methods outlined in this chapter. One of the greatest challenges for psychology has been to
explore and describe the human mind when it is not, in itself, a thing that can be directly
observed. In recent times, technology has helped us observe the brain in action, and this has
unquestionably aided psychologists in their attempts to be empirical.
Following the scientific method requires both an empirical approach and an attempt to be
as objective as possible. Being objective requires the researcher to be unbiased. This means
that his or her thoughts about the topic should have no influence on the evidence gathered
or the interpretation of it. In this way, we hope that the conclusions reached by researchers
are trustworthy; that if others did the same research, they would observe similar results; and
that they would draw similar conclusions.
Operational definitions are
descriptions of variables
that are specific and
In order to achieve this, good scientific research needs to have several characteristics. First,
the topics being studied – the behaviours or the aspects of mind under investigation –
must be clearly defined. Researchers use operational definitions to explain what they are
measuring. For example, it is not enough to say that you are measuring ‘aggression’ in
children; it is necessary to have an explanation of specific aggressive behaviours that will be
When we understand exactly what is being measured, we also need to be convinced of two
things: that the data obtained by the researcher is reliable and that the method used was a
valid way to obtain the data.
Reliability (similar to accuracy) can be established by considering the following questions.
1 Did more than one person record and interpret the data, and do they agree?
The answer to this question relates to the idea of inter-rater reliability. This can be very
important in psychological research where people might disagree about whether the
behaviour they have observed, even in a highly controlled environment, does in fact
constitute an occurrence of the behaviour they are interested in recording. For example, if
two people were recording the number of laughs made by an individual while watching a
television comedy, they might easily disagree.
How easy is it to measure
how much people laugh?
2 If you use the method again in the same situation, do you get the same results?
An acceptable answer to this question depends on two things. First, that the research can
be repeated. This is referred to as the replicability of a study and is, in turn, dependent
on the researcher giving clear details of the method used, and in particular providing
operational definitions. Secondly, ideally the study has been replicated and similar
results found. If we use a reliable test with the same participant, for example, we should
get the same result. This is called test–retest reliability.
Ideally, all research needs
to be replicable so that
findings can be confirmed
Validity can seem a complex idea. It can be broken down into two types: internal validity
and external validity. Internal validity is concerned with the quality of the research itself,
particularly in experiments when the researcher wants to make cause–effect claims.
Researchers should be studying what they claim to be studying, and measuring what
they claim to be measuring. Is counting a viewer’s laughter a valid way of measuring how
funny a television comedy is? The independent and dependent variables specified by
the researcher must be clearly defined and be a fair reflection of the phenomenon being
investigated. When the study is carried out, extraneous variables such as noise or changes
in behaviour because of the experimental situation should be eliminated or their effects
minimized. External validity is concerned with how appropriate it is to apply the results of
the study to the intended population.
Again, there are several questions we should consider before deciding whether researchers’
methodology and conclusions are valid or not. Some of the most important questions are
examined in the following pages.
Internal validity issues
1 What is the researcher trying to manipulate (in experiments) and measure? Is this really
what they measured? Does it seem to be what they are measuring?
Research has face
validity if it appears
to be investigating
the phenomenon the
to examine. A lot
of intelligence and
uses tasks that measure
something but it is not
clear that they measure
This kind of validity is probably the most important, and it is another reason that
operational definitions of variables are so important. Many psychologists argue that IQ
tests, for example, are measuring something, but not exactly what people mean when
they are talking about intelligence. If research at a very superficial level seems to have
been done well, then it can be said to have good face validity.
2 Did the location or nature of the research somehow make the participants act in a
Sometimes participants try to guess the nature of the research they are participating in
and then act accordingly, which is one reason why psychologists sometimes prefer to
keep it secret. Demand characteristics are the effects that occur because of this. The
Hawthorne effect occurs when participants try to perform in a way that they think
meets the expectations of the researcher. A lesser known opposite of this is the screwyou effect, where participants act in a way that might sabotage the researcher’s aims.
If research is done in a more natural setting, or if participants know little about the
research or their participation in it, demand characteristics are less of a problem.
Children often try hard to
please the experimenter.
3 Has the researcher maintained objectivity in their interpretation of the results?
Reading the conclusions of researchers after reading through their results can
occasionally be confusing or even disappointing. Sometimes it is clear that the researcher
wanted to be able to draw certain conclusions, and the data have been misinterpreted,
variables poorly defined, or the researcher has relied only on their own judgement
through the whole process of research. It is particularly difficult to maintain objectivity
when researchers have hopes and expectations about the conclusions they might be able
to draw from their research.
External validity issues
1 Was the research done in an artificial environment, or were the tasks performed
Sometimes this is referred to as the artificiality of a study, at other times it might be
referred to as the ecological validity. The latter term focuses mostly on the environment
in which the work was carried out. Experiments in laboratories enable the researcher to
control the conditions and eliminate many variables that might interfere with his or her
work. However, this can create an environment so unrealistic that it becomes impossible
to generalize from the artificial environment to the real-life environment to which the
results should apply. Also, when researchers attempt to measure psychological processes,
they often ask participants to perform tasks that are not like their normal activities.
Brain-imaging technology, for example, is giving us a better understanding of how the
brain works, but researchers are very limited in what kind of activities they can ask
participants to perform as they must often lie inside a large machine.
Many experiments lack
ecological validity because
they are carried out in
an unnaturally pure
environment, free from
the influence of unwanted
variables and need to
be replicated in a more
natural setting before we
can consider the findings
It is difficult to move inside
an MRI machine, which limits
how much we can learn from
2 Can we generalize from the participants in the sample to the wider population?
There are two very simple examples that illustrate this point. First, a lot of research
has been done on animals because it is considered unethical to do the same research
on humans. One of the reasons it has been possible to do this kind of research,
is the assertion that animals and humans are fundamentally different in terms of
consciousness or ability to feel pain. If it is accepted that animals are different from
humans in this way and therefore it is ethically acceptable to do the research, is it valid to
generalize from animals to humans?
The second example concerns sampling in psychological research carried out at
universities. An estimated 75% of research has used undergraduate students as
participants, with psychology students making up around a third of all research samples
(McCray et al., 2005). Are psychology students representative of the general population?
In some ways they are, but it is usually better to find a more varied sample if you plan to
There are a number of other factors that can render the conclusions a researcher draws
invalid, and examples of these will be addressed when you look at individual studies and
Whenever you are asked to
evaluate, discuss or compare
research, you can start by
considering reliability and
validity in general. Each
method has typical strengths
and weaknesses and these are
discussed in this section.
What is good research? General ethical
Good research needs to be ethical for several reasons. One reason is to preserve the
reputation of psychology in the academic world and among the public – particularly to
ensure that people will continue to come forward to participate in research. There are
several important historical examples of research that was done without the consent of
participants, or that caused significant psychological or physical harm. In the USA and
later in the UK, guidelines were drawn up to try to ensure that all research done within the
discipline is not damaging to institutions carrying out research or to the individuals who
Guidelines for research involving people
The following are key points to understand from the British Psychological Society (BPS)
and the American Psychological Association (APA).
• Approval is gained from the institution the researcher is working for. This is relevant
for IB students in that before you carry out research for your internal assessment, you
need a full understanding of the ethical requirements and you should not proceed
without the approval of your teacher.
• Informed consent is obtained from all those who will participate. Consent indicates
agreement. Informed means that participants have been made aware of the purpose,
duration and procedures of the research; and of their rights, benefits and any possible
negative consequences of participation. Of course, sometimes it is not possible to give
full information about the research, and it is acceptable not to obtain informed consent
if no harm is expected and only anonymous questionnaires or naturalistic observation
are used as methods. The BPS also stresses that records of informed consent should
be kept and that naturalistic observations should be done without consent only if
the behaviour observed would normally be expected to be observed by strangers.
This is usually understood to mean that it must occur in a public place. Additionally,
researchers should take cultural considerations into account in deciding what behaviours
might be acceptably observed or recorded. The BPS also specifies that children under the
age of 16 years, and adults not competent to understand the nature and purpose of the
research should not be allowed to consent alone: a guardian or family member should
• Deception is to be avoided unless it is justified by the potential significant contribution
that the research will make, and the impossibility of avoiding deception. If psychological
harm or physical pain are expected, deception is explicitly not allowed. When
participants are deceived, they should be informed about this at the earliest possible
opportunity. Clarifications in the BPS guidelines emphasize that there is a difference
between falsely informing participants about the nature of the study and the more
ethical withholding of information that might affect the participant’s behaviour while
being researched. Researchers should use their judgement to avoid doing research if
participants would be upset when they find out the true nature of the research.
• Debriefing must take place. This means that the nature, results and conclusions of
research need to be made available to participants as soon as possible. Any harm the
participant experiences and of which the researcher becomes aware is then minimized.
In addition, where research has effected a change in physical or psychological state,
these effects are undone. The BPS refers, for example, to research that induces a negative
mood state. It would be wrong to cause a bad mood in participants and allow them to
leave still in a bad mood.
• Participants’ rights include the right to confidentiality and the right to withdraw. The
right to withdraw is usually upheld at all points during the research process, including
the point after data has been collected and the participant has been debriefed. In
addition, the BPS guidelines stress that if any inducements to participate were offered,
such as academic credit or payment, these should not be affected by a participant’s
decision to withdraw.
• Fabrication of data is unacceptable, and any errors should be later corrected.
Guidelines for research involving animals
Both the BPS and the APA provide guidelines for doing research with animals. There is a
great deal of controversy surrounding the use of animals for research in a number of fields,
and it is important that guidelines are followed to prevent clearly unjustified harm or illtreatment to animals. The guidelines suggest the following key points.
Is animal research ethical?
• There must be a clear scientific purpose that will increase knowledge of processes
relating to the evolution, development, maintenance, alteration, control or biological
significance of behaviour; the research should also benefit the health or welfare of
humans or other animals.
• If the procedure would cause pain in humans, it should be assumed that it will also
cause pain in animals, and should therefore have very strong justification. If any aversive
stimuli are used, these should be set at the lowest possible levels. The APA recommends
that psychologists test all painful stimuli on themselves.
• Animal welfare should be monitored throughout the research. Animals should be treated
humanely inside and outside the context of research.
• When serious or long-term harm is caused to animals, they should be euthanized
(killed) as soon as possible. Although this may seem harsh, it is considered the ethical
thing to do if the research has caused damage to the animal that will affect its ability to
live a relatively normal, pain-free life.
To access Worksheet
1.1 on evaluation of
research, please visit www.
and follow the on-screen
Animals used in research
laboratories must be treated
Personal space invasion (Middlemist et al. 1976)
This research took place in a men’s toilet in which men normally had free choice about which
of three individual urinals they used. In order to test the effects of having another person close
to a man while he urinates, the researchers randomly assigned visitors to the toilet to one of
three conditions. Participants entering the first condition were forced to use the urinal closest
to the toilet stalls because the furthest urinal bore a sign saying ‘Don’t use, washing urinal’, while
a confederate occupied the middle urinal, pretending to be urinating. In the second condition,
the confederate and the sign were switched so that the out-of-order urinal was between them,
increasing the personal space between the men. The third condition involved no confederate.
Most of the research in
this book is considered to
be ethical, but you should
consider what ethical
problems researchers might
have had to overcome and
how breaches of ethical
guidelines might have been
A student measured onset and persistence of urination in the unknowing participant using a
stopwatch and a periscopic prism that allowed him to witness the signs that it had begun from
inside the toilet stall next to the urinal, but did not allow him to see the participant’s face. The
researchers found that having a man standing at the nearest urinal increased the time taken for
urination to begin and decreased its persistence. Participants were not informed that they had
been involved in an experiment.
The research study above has been criticized on ethical grounds. Evaluate it in terms of
validity, reliability and ethics.
In order for researchers to be able to state with complete confidence that they have accurate
results that apply to the entire human population, the research would need to be carried
out on the entire human population. This is clearly impossible, and therefore researchers
take a sample of the population under investigation that they expect to be representative of
sampling is the ideal, it
is not often possible in
In order to be sure that they have a representative sample, several considerations need to
be taken into account. First, a target population needs to be identified. While researchers
often hope the target population would include all humans, it is more likely that they will
acknowledge that differences such as culture and gender mean that they cannot apply their
results to everyone. It is more appropriate, therefore, that the target population is more
narrowly defined. Such a population could be, for example, all registered users of a website,
or students from your school, or customers of a telecommunications company. Ideally,
random sampling is then done from the target population in order to avoid sampling
bias. For example, research carried out on students at the beginning of a university lecture
will fail to include those students who arrive late for class. This is important if it can be
demonstrated that there are differences between those who are included in the sample and
those who were excluded. Some techniques commonly used by researchers to obtain their
sample are outlined below.
1 Random sampling – If a list of all members of the target population is available, the
researcher can then ensure that every member has an equal chance of being invited to
participate, for example by numbering members and selecting them using a random
number generator, or by pulling names out of a hat. Clearly, the more people in the
target population, the more time-consuming this will be. In addition, there is still the
risk that, because not all people who are invited to participate will agree, there could be
a bias in the sample if there are important differences between those who are willing