Methods
Self- Report Data – Questionaires and Interviews
A survey is a self-report method designed to Quantitative Data – information that is
gather large amounts of information through a numerical or can be converted into numbers
questionnaire or large-scale interview.
Qualitative Data – non – numerical
Questionnaires information
• Designed to gather large amounts of data
by accessing a large sample. Questionnaires can produce both kinds of
data.
• Administered face to face, online, by post,
by email.
• Require info from participants about their
attitudes, opinions, lifestyles etc
Closed questions
Closed questions – preset fixed answers that Attitude scales involve more than yes/no and
respondents selects by circling or ticking the give a range of options so that their strength
answer closest to their opinion. can be assessed.
Produce quantitative data. A Likert-scale involves a respondent selecting
from a fixed set of options to rate their
Can be yes/no responses or a list of options agreement to a series of statements.
E.g. A Likert scale was used in Adorno et al. survey
on authoritarianism.
+ quick and easy to answer
+ easy to analyse
-limited information
- Do not know why the respondent answered
in that way (limited detail)
Rating vs Ranking
What is a Rating Scale? What is a Ranking Scale?
Rating scales require people to pick a value to A ranking scale forces respondents to rank a list of
describe their level of satisfaction or agreement items with only one selection in each rank. Ranking
with something. Questions Examples
Example 1: Rank the following snow cone flavors
Examples of Rating Scale Questions
from 1 to 5, with 1 being “keep them coming” and 5
NPS (net promoter score) being the “no way.”
On a scale of one to ten, how likely are you to •Grape
recommend (your brand here) to a friend or •Cherry
colleague? •Lemon
very unlikely – 1 2 3 4 5 6 7 8 9 10 – very likely •Strawberry
•Orange
Likert Scale Example 2: Rank the following items by how much
Rate your level of agreement with the following they influence your choice to buy a car, from that’s
it (1), not so much (2), and don’t care (3).
statements:
•Price
I decide to buy a car based on the price. •Gas Mileage
Strongly agree | agree | neither agree nor •Cargo Space
disagree | disagree | strongly disagree
Rating vs Ranking
Limitations of Rating Scale Questions
Limitations of Ranking Scales
The biggest drawback to ranking scale
Not as helpful with longer lists of items - the
questions is a phenomenon called satisficing.
longer your list, the more likely respondents
People don’t put effort into evaluating your
are only to have extreme opinions about the
questions and give everything the same rank
best and worst options. You probably can’t
because it’s easier.
gain much insight from what people put in the
middle tier.
It will reveal what most people prefer, it will
not explain why they have that preference.
If people like some things equally, they may
resent being forced to rank them.
Objective and Subjective
ACTIVITY : Practise questions to do in your books
Describe one strength of questionnaires for use in social psychology. (2 marks)
Describe one weakness of questionnaires for use in social psychology . (2 marks)
Imagine you are designing a questionnaire for Year 11 to investigate the popularity of different A
level subjects at school to support Jonathan in deciding which A levels to offer next year.
Design a questionnaire using as many of the different types of questions as you can (closed
questions, attitude scales, likert scales, open questions). You can do this using google forms if you
prefer. Work with a partner who is confident using google forms. Share the questionnaire with
Abby.
Issues with questionnaires – Social Desirability
Social Desirability
Social desirability occurs when a
respondent does not give a
genuine answer, but one that
reflects them in a more
favourable light i.e. more
conforming with social norms.
E.g. very few people would say
on a questionnaire that they
agree with cheating in exams
because this response is contrary
to social norms.
This can be problematic for
research that investigates
sensitive issues or attitudes that
go against social norms.
Issues with questionnaires – Question Construction
• Important to ask questions that are not too ambiguous, technical or complex.
• Important to ask questions that do not mislead a respondent.
• Important to ask questions that do not violate the respondents right to privacy.
• Important to consider the number of options when designing a Likert scale because an odd
number of possible answers may mean the middle value is chosen more frequently. An even
number of responses forces the respondent to make a choice and not remain neutral.
• Important to mix up and reverse statements to avoid response bias where the respondent
may just slip into agreeing or disagreeing with the statements.
Issues with questionnaires – Questionnaire reliability
Reliability refers to the consistency of a measure or a finding.
External reliability refers to the consistency of a measure or finding over
time.
Some questionnaires loses their external reliability if respondents repeat
them on different occasions so it is important to establish whether this is
the case by using the test –retest method.
This literally means that the same people are given the same
questionnaire to complete again on a different occasion. If their
responses are the same or very similar, external reliability can be
established.
Issues with questionnaires – Questionnaire reliability
Reliability refers to the consistency of a measure or a finding.
Internal reliability refers to the consistency of a measure within itself.
Internal reliability is a problem for questionnaires because oftenseveral
different questions are used to measure the same trait or attitude. To
establish internal reliability, the split-half method can be used. This
involves splitting all the questions into two halves and comparing the
findings from both halves. If all the questions are measuring the same
concept, both halves should achieve the same score.
Issues with questionnaires – Questionnaire validity
Validity refers to the extent to which something is measuring what it intends to
measure and nothing else.
Sometimes this can be established by looking at each question in turn and deciding
whether it makes sense in terms of the concept/idea being measured. This is known as
Face validity.
Validity can be confirmed by asking an expert in the field to review the questions.
If a questionnaire is a valid measure of something e.g. intelligence then it should have
predictive validity meaning it can accurately predict the same thing in the future.
E.g. if an intelligence test has predictive validity, high intelligence scores on the
questionnaire should correlate with education success e.g. good A Level grades.
Another way of establishing whether the questions in a questionnaire are valid is to
compare it to another test measuring the same thing – this is known as concurrent
validity.
Write these terms in your glossary
RELIABILITY VALIDITY
External reliability Face validity
Internal reliability Predictive validity
Test-retest method Concurrent validity
Split half method
Interviews
An interview is where a respondent answers questions
directly, rather than writing them.
This could be a conversation or simply answering questions
directly from the interviewer.
Interviews can be structured, semi-structured or unstructured.
Structured Interviews
Structured interviews are defined by the nature of the questions and the way in which they are
asked.
Typically structured interviews are standardised so that all respondents are asked the questions
in the same way.
+ easy to administer
+not necessary to establish rapport between researcher and respondent
+easily repeated
- They can be superficial and lack depth
- Respondent may not be able to express their opinions fully
Semi structured interviews
Semi-structured interviews are more conversational and dynamic. The researcher has
planned questions that they aim to be answered, but do not have a standardised
format to follow and can deviate from the script.
+ conversation can flow better
+ more information obtained
+gathers both qualitative and quantitative data
Unstructured Interviews
These begin with a loose research aim and gather information from respondents in the form of a
conversation.
Unlike structured interviews the researcher needs to be analytical during the interview so that
they can probe and seek meaning from respondents.
Unstructured interviewers need to be skilled in achieving good rapport with respondents and be
responsive to the information offered.
They need good listening skills.
They need to be non-judgemental in their language.
- More time consuming to conduct
- More time consuming to analyse data
- Not replicable as different questions are asked in each interview
ACTIVITY
Identify which type of interview is being described:
1. A researcher wants to investigate how friendships are formed. They decide to write specific
questions that they will ask people in friendship groups.
2. A researcher has a conversation with students in the dining room to find out which meals
they prefer.
3. A researcher wants to investigate what pets people own. They want to know what type of
pets people have and how long they have had them but want to explore the reasons for
owning pets.
4. Show some example questionnaires and sort
Ethics
Ethical issues are important when conducting any type of questionnaire but especially
unstructured interviews because the qualitative data gathered may make direct
reference to quotes from respondents.
It is important that the respondents details are anonymised and personal details
disguised.
The interviewer must behave sensitively when asking the respondent for personal
information to ensure they do not breech their right to privacy.
Researcher Effects
Interview characteristics
The sex, age, manner and personality of the interviewer can affect how a persona responds and
whether they are truthful or disclose any information at all.
Important to predict what characteristics might influence the respondents and control them. E.g.
you can predict that a male interviewer is unlikely to obtain detailed information from female
participants about friendships. This can be controlled by employing a female interviewer.
Confirmation bias
A researcher may inadvertently ask a question tat serves to confirm their expected outcome of
the study.
This can be reduced by employing an independent researcher who is trained in interview
techniques but does not know the aim of the study.
Activity
Exams question may ask you to give a strength or weakness about
interviews used in social psychology.
E.g. One strength of using interviews in social psychology is that depth
answers contain detailed information and reasons why individuals may
consider others to have authority or power. This increases the validity of
the data gathered by social psychology researchers about topics such as
obedience.
Describe a weakness in your book….
Primary and Secondary Data
Primary Data is gathered directly by a researcher from participants or sources typically using a
research method such as an experiment, observation, questionnaire or interview.
Primary data is first hand data and therefore typically gathered according to the aims of the
study.
Primary data is time consuming to collect as the researcher must gather the participants and
design and conduct the research themselves.
Primary data is current.
Secondary data is gathered from existing studies conducted by other researches, such as when
conducting a literature review or meta-analysis. Large amounts of data can be gathered from
many studies.
Secondary data is second hand data and may be slightly different and not specific to the aims of
the researcher.
Secondary data has already been collected so just needs to be analysed and compared and
therefore less time consuming.
Secondary data may be out of date.
Alternative Hypotheses
In addition to the overall aim of the research, a study also makes a prediction about what is likely
to occur. This prediction is known as an alternative hypothesis.
A hypothesis should contain the variables under investigation and be a clear, testable and
precise statement at the beginning of the study.
The predication is often guided by previous research in the topic area, but if there is limited
research or mixed findings, the prediction may have to just state that a difference or a
relationship might be found between the variables under investigation but not what direction the
difference/relationship might take.
Sampling Techniques
In psychological research it is necessary to recruit participants or respondents to study. The way
in which these participants are selected is known as sampling.
It is unlikely that the whole population can be studied, so a sample population needs to be
gathered using a sampling technique.
The aim of the sampling technique is to select a representative sample of participants – a sample
that represents the characteristics of the population well.
The technique used will depend on the type of research being conducted and the availability of
the participants.
Target Population
A target population is the population of people being investigated by a study.
e.g. if you are investigating attitudes about health services at the local hospital then the target
population will be people at the hospital. A sample will be recruited from this target population
using a sampling technique.
If the sample gathered is not representative of the target population because of an under or over
representation of a particular type of participant then a sampling bias will occur.
Stratified Sampling
If the target population has certain subgroups that need to be represented in the sample then a
stratified sampling technique can be used to recruit the correct number of participants in those
sub groups.
The target population needs to be divided into those subgroups and the percentage of members
of each subgroup should be calculated.
Stratified sample can be useful if subgroups are unequal or need to be represented in a sample.
E.g. if you are researching stress in the workplace you can find out how many staff occupy
different roles within the company. E.g. if there are office clerks, managers, canteen staff,
cleaners. There may be more clerks that managers and so more clerks need to be recruited for
the study than managers to represent the company staff more fairly.