Psychoogy Paper 3

Qualitative- why? 

Can provide a deep, detailed understanding of why people think, feel, and act in certain ways. However, they also have several major drawbacks such as being more time-consuming and costly than quantitative studies.

Quantitative- objective, how often/much?

Provides clear, measurable answers, reduces bias through statistical analysis, and allows findings to be generalised and replicated

Case studies 

Naturalistic observations 

• participants/ non participants 

• covert/ overt 

Interviews

• Unstructured

• Semistructured 

• Focus group interviews

Experiments 

• Field 

• Natural 

• Quasi 

• Lab 

Correlational research

Surveys 

Research Method

What is it? 

Advantages 

Disadvantages 

Field Experiment 

Setting - natural real word environment.

IV - deliberately manipulated

DV - measured

Extraneous variables - minimal control

→ High ecological validity and mundane realism 

→ Low demand characteristics 

→ Low validity 

→ Low reliability 

Natural Experiment 

Setting - any setting

IV - not manipulated, but naturally occurring (event or experience)

DV - measured

Extraneous variables - little to no control

→ High ecological validity 

→ High mundane realism 

→ Low internal validity 

→ Low reliability 

Quasi Experiment 

Setting - any setting

IV - pre existing characteristics, so not manipulated

DV - measured

Extraneous variables - little to no control

→ High ecological validity 

→ High mundane realism 

→ Low internal validity 

→ Low reliability 

Lab Experiment 

(True Experiment) 

Setting - a tightly controlled artificial experiment.

IV - deliberately manipulated by the experimenters.

DV - the experimenters measure it. It will produce quantitative data.

Extraneous variables - attempt to control and minimise them.

→ High internal validity 

→ High reliability 

→ Low ecological validity and mundane realism 

→ High demand characteristics 

Correlational Research 

Measures the extent of which two variables are related. Researchers do not manipulate variables but only observe.

→ Allow researchers to measure and study variables that are occurring in large populations. 

Correlation does not mean causation, if two variables are related it does not necessarily mean that one variable is causing the other.

Researchers have less controls over external variables which could affect the variability and the reliability of the findings.  

Surveys 

Standardised questions (close ended) 

Large sample 

Analysed statistically to find patterns or correlations 

→ Efficient data collection as allowed of variety of question types such as open ended questions, which increase the richness of data 

→ Allows for statistical analysis and generalisation to a population

→ Able to reach large sample sizes 

→ Low effort and high efficiency 

→ Standardised 

→ Lacks depth and detail 

→ Low eternal validity 

→ Risk of response bias

Research Method

What is it? 

Advantages 

Disadvantages 

Case Studies 

A type of research method that involves using multiple methods to gather lots of data on a small group or individual.

• It is a naturally occurring phenomenon and is not manipulated by the researcher

• May be conducted over a long period of time, and/or

uses a range of different methods to gather lots of data

→ In depth data can be qualitative or quantitative 

→ High ecological validity 

→ Is able to study change over time

→ Triangulation leads to a higher level of validity 

→ Cannot be replicated 

→ Often retrospective, lacks data on the individual before the accident/event/start of symptoms 

→ Time consuming and requires good training of a team 

Naturalistic Observations 

Involves collecting data from observing participants in their natural environments without manipulation of the setting. 

Researchers can observe participants' naturally occurring behaviour as a result of high ecological validity and mundane realism. This enhances the applicability of the findings to real world contexts.  

Data collected heavily relies on researcher interpretation which could be subject to researcher bias, decreasing validity and reliability of findings.

There is also an inability to account for and control extraneous variables or manipulate an IV, resulting in an inability to establish causation.  

Participant Observation 

The researcher becomes a member of the group they are observing

→ Researcher gains familiarity with the given area of interest; insider perspective

→ Minimises participant bias, as behaviour is naturalistic

→ Observation is more insightful; some findings may not be discovered otherwise

→ Breach in ethicality - involves deception; the observed participants have not consented

→ Low replicability - observed pts may behave differently depending on researcher; findings not reproducible (observer effect)

Non-Participant Observation 

Researchers remain separate and do not interact with the participants whilst observing their behaviour. 

→ Less intrusive 

→ Reduces demand characteristics 

→ Highly controlled over variables 

→ Demonstrates cause and effect between observation 

→ May miss contextual details 

→ Artificial setting may reduce ecological validity 

→ Ethical concerns about the observation 

Covert Observation 

Participants are unaware that they are being studied, where it is conducted in a natural setting, and their behaviour is being observed. (Researcher identity and purpose of the experiment are hidden from the participants)

→ Demand characteristics are prevented, since they are unaware that they are being watched. 

→ High ecological validity 

→ It's difficult to record data openly without risking exposure

→ It's unethical because it removes participants’ right to withdraw 

→ Its difficult to replicate the study, due to uncontrolled natural setting 

Overt Observation 

Subjects are aware they are being observed. 

The researcher’s identity as an observer is not hidden from the participants. Conducted in a natural environment. 

A researcher can ask participants direct questions to clarify their actions and intentions; therefore, the results are more accurate. 

→ High chance of demand characteristics 

Unstructured Interview 

Interviews that have no fixed format, open ended questions like a conversation. Questions may depend on the interviewees responses. 

→ In depth qualitative information and insight to personal experiences/ perspectives. 

→ Useful for researching complex topics. 

→ May have bias from interviewers' questions and interpretation of responses. 

→ No standardisation.

→ Time consuming and requires a skilled interviewer. 

Semi-Structured Interview 

It has pre-set questions that are prepared beforehand and a framework to follow, but also some open-ended questions or points of discussion that the researcher can follow up with the respondent

→ Flexible open ended questions 

Giving respondents the freedom to answer in their own way 

→ More time consuming than structured interviews 

→ Interviewers require more training on how to conduct an interview 

Focus Group Interview 

• Small group discussion based on a topic given 

• Observations 

• Videoing 

• Used to identify and explore how people think and or behave

• Open conversation 

→ Rich qualitative data

→ Diverse perspectives, therefore range of viewpoints

→ Group interaction can stimulate discussion/new ideas

→ Can adapt to flow of conversation 

→ Participants can conform to dominating opinions

→ Effectiveness of a focus group depends heavily on the moderators skills as a poorly guided discussion can lead to irrelevant data

Sample: 

• A small number of people taken from the target population who participants in the investigation 

• As representative of the target population as possible so that the psychologist can generalise the results to the target population without having to study all of them 

• A representative sample should have the same characteristics as the target population 

Sampling Bias - A sampling bias has occurred if the target sample selected under or over represent certain groups that compose the target population

• In this case, the sample is unrepresentative of the target population 

• This limits the extent to which generalisations can be made

• To avoid this, the sample should be as large as possible

Sampling Method

Details 

Advantages 

Disadvantages 

Random Sampling 

(probability) 

Where every participant in the target population has an equal chance of being selected 

• Through a lottery systems, with no bids from the psychologist 

• Those selected are then contacted and inhibited to participate 

The sample is likely to be unbiased representative of the target population. This is because everyone in the target population has the same and equal chance of being selected without bias from the researcher

• This means the findings from the study can be generalised back to the target population

It is more time consuming and inconvenient compared to opportunity sampling

• This is because the researcher has to obtain information about everyone in the target population to give each one an equal chance of being selected 

• This is often impractical if the target population is large

• This means that random sampling is difficult to achieve in psychological research because it is time consuming 

Opportunity Sampling/ 

Convenience Sampling   

(non-

probability)

This involves approaching and inviting those who are available, ie. Who happens to be present at the time and place the researcher is looking for participants

• e.g. asking students in a university library to take part in a study conducted at the university

This is more convenient to use as it is less time consuming to obtain as the psychologist can use anybody who is available or willing at the time, and there is no need to gather information about the whole target population 

• Which means opportunity samples often used in psychological research as they save a lot of effort 

The sample of participants is unlikely to be representative of the target population because the sample is more likely to be biased (sampling bias) as participants often share similar characteristics given they are selected from one place at one time 

• This means it is difficult to generalise the findings back to the target population 

Volunteer Sampling

(self selected) 

The sample is self selected. This involves advertising the study and providing contact details, so that individuals respond if they wish to participate in the study.

• e.g. placing an advertisement in a local newspaper that asks for volunteers to telephone the number provided if they want to take part

More convenient as they are less time consuming to obtain because individuals approach the research themselves, the researcher does not need to seek them 

The sample is unlikely to be representative of the target population as the sample is more likely to be biased as volunteers have similar characteristics.

This means that it is also difficult to generalise the findings back to the target population. 

Purposive Sampling 

(non- 

probability)

Has inclusion or exclusion criteria, i.e. only participants with/without certain characteristics can take part in study

• Selecting sample based on characteristics of population and the objective of the study

• Not random

→ Flexible 

→ Cost efficient 

→ Suitable for small or hard-to-reach populations

→ Targets specific participants

→ Not representative beyond the target population 

→ Not generalisable 

→ Potential researcher bias as not random sampling 

Snowball Sampling 

Participant recruits more people also relevant to the study (asked to refer others they know relevant to target population about participation in the study)

• Not random

→ Easy way to obtain participants

→ Suitable for hard-to-reach populations

→ Lower costs 

However, participants will come from the same social circle, meaning the sample may not represent the entire target population

→ Unrepresentative sample 

→ Lack of generalizability 

Informed consent 

Participation in a study must be voluntary, and participants must fully understand the nature of their involvement, including the aims of the study, what tasks they will be exposed to and how the data will be used. Researchers should provide as much information as possible and in the clearest possible way, hence the name “informed” consent. If the participant is a minor, consent should be obtained from parents or legal guardians.

Protection from harm 

At all times during the study participants must be protected from physical and mental harm. This includes possible negative long-term consequences of participating in a research study.

Anonymity 

Guarantees that participants cannot be identified

Confidentiality 

Ensures that the researchers protect the participants' identities and their responses

Withdrawal from participants 

It must be made explicitly clear to participants that, since their participation is voluntary, they are free to withdraw from the study at any time they want. Researchers must not prevent participants from withdrawing or try to convince them to stay. 

Deception 

It should be kept to a minimum; the true aim of the study will not be revealed towards the participants to prevent demand characteristics from the participants. 

Debriefing

After the study participants must be fully informed about its nature, its true aims, how the data will be used and stored. They must be given an opportunity to review their results and withdraw the data if they want to. If deception was used, it must be revealed. Care must be taken to protect participants from any possible harm including long-term effects such as recurring uncomfortable thoughts. In some cases psychological help must be offered to monitor the psychological state of the participant for some time after the study (for example, in sleep deprivation studies).

Truthful 

Participants should be fully informed about the aim and procedures of the study

Careful 

If it is a sensitive issue, it could potentially stress the participant and make sure participants weren’t psychologically harmed. 

Data fabrication 

• This is a serious violation of ethical standards and psychologists may lose their license if they fabricate data. If an error is found in already published results, reasonable measures should be taken to correct it. 

• Researchers should also be truthful when reporting data, this is called integrity, which includes the principles of honesty and accuracy. This means researchers must avoid fabricating, falsifying, or misrepresenting data and instead report it accurately and objectively. 

Plagiarism 

It is unethical to present parts of another’s work or data as one’s own.

Publication Credit 

Authorship on a publication should accurately reflect the relative contributions of all the authors.

For example, the APA Code of Ethics states specifically that if a publication is based primarily on a student’s work, the student must be listed as the first author, even though his or her professors co-authored the publication.

Sharing research data for verification 

Researchers should not withhold the data used to derive conclusions presented in the publication. This is encouraged for transparency and reproducibility, but must be done ethically by prioritizing participant privacy and confidentiality. 

Generalisability (quantative) - are the results applicable to the whole population? 

• Sampling bias 

• Sampling technique 

• Sampling size 

• Nature of the task 

• Ecological validity  

Transferability (qualitative) - can the findings be transferred to other settings or populations? 

• Representational 

• Inferential 

• Theoretical 

Type of Generalisation 

Explanation 

Equivalent in quantitative research 

Sample-to-population generalisation 

Applying the results of the study to a wider population. It depends on how representative the sample is. The best way to ensure representativeness is to sample randomly.

However, since the nature of sampling in qualitative

research is non-probabilistic, this type of generalization in qualitative research is a weak point.

Population validity 

Theoretical generalisation 

Generalizing results of particular observations to a broader theory. Theory plays a much greater role in qualitative research. We can generalize to a broader theory if data saturation has been achieved. Data saturation is a point when further data does not add anything new to the already formulated conclusions and interpretations.

A similar idea is construct validity because it refers to the ‘leap’ from observable construct. 

Case-to-case generalisation (transferability) 

Applying the findings of a study to a different group of people or a different setting or context. In qualitative research case-to-case generalization is the responsibility of both the researcher and the reader of the research report. The researcher ensures that thick descriptions are provided

so that the reader has sufficient information. The reader decides whether or not the new context is similar enough to the one described in the report for the findings to be applicable.

Ecological validity (generalising from experimental settings to real life settings)

Validity (quantitative) - does the research measure what it intends to do

Internal validity - high levels of control within the experiment

• Standardised instruction 

• Controlling confounding variables 

• Operationalising variables 

External validity - conclusions hold up beyond the setting

• How realistic was the situation 

• Remember high levels of internal validity 

Credibility (qualitative) - is the extent to which results of the study can be trusted to reflect the reality 

• Triangulation - combining and comparing different approaches to collecting and interpreting data:

  • Method triangulation—combining different methods

  • Data triangulation—using data from a variety of accessible sources

  • Researcher triangulation—combining and comparing observations of different researchers

  • Theory triangulation—using multiple perspectives or theories to interpret the data.

• Establishing a rapport - Building a relationship of trust with the participant, emphasizing the necessity to be honest, the right to withdraw and the fact that there are no good or bad answers. All this prevents participants from altering their behaviour in the presence of the researcher.

• Iterative questioning - Returning to the topic later in the process of interaction with the participant and rephrasing the question. This allows a deeper investigation of sensitive topics.

• Reflectivity 

• Credibility checks - Checking accuracy of data by asking participants themselves to read transcripts of interviews or field notes of observations and confirm that the notes/transcripts reflect correctly what the participants said or did.

• Thick descriptions - Describing the observed behaviour in sufficient detail so that it can be understood holistically and in context. Contextual details should be sufficient to make the description meaningful to an outsider who never observed this behaviour first-hand.

Bias 

Explanation 

Ways to overcome the bias 

Acquiescence bias 

A tendency to give positive answers whatever the question. It may occur due to the participant’s natural agreeableness or because the participant feels uncomfortable disagreeing with something in the research situation. 

Researchers should be careful not to ask leading questions. Questions should be open ended and neutral. It should be clear that there are no ‘right’ or ‘wrong’ answers. 

Social desirability bias 

Participants’ tendency to respond or behave in a way that they think will make them more liked or more

accepted. Intentionally or unintentionally, participants may be trying to produce a certain impression instead

of behaving naturally, and this is especially true for

sensitive topics.

Questions should be phrased in a non-judgmental way. Good rapport should be established. Questions can

be asked about a third person

Dominant respondent bias 

Occurs in a group interview setting when one of the

participants influence the behaviour of the others.

Other participants may be intimidated by such people

or feel like they will be compared to the dominant

respondent.

Researchers should be trained to keep dominant respondents in check and try to provide everyone with equal opportunities to speak.

Sensitivity bias 

A tendency of participants to answer regular questions

honestly but distort their responses to questions on

sensitive topics.

Building a good rapport and creating trust. Reinforcing ethical considerations such as confidentiality. Increasing

the sensitivity of the questions gradually.

Confirmation bias 

Occurs when the researcher has a prior belief and uses

the research (intentionally or unintentionally) to confirm

this belief. It may manifest itself in such things as

selectivity of attention or tiny differences in non-verbal

behaviour that may influence the participants.

Strictly speaking, this is unavoidable because in qualitative research the human observer is an integral part of the process. However, this bias can be recognised and taken into account through the process of reflexivity.

Leading question bias 

Occurs when the questions in an interview are worded

in a way that encourages a certain answer. For example,

“When did you last have angry thoughts about your

classmates?”

Interviewers should be trained in asking open-ended, neutral questions.

Question order bias 

Occurs when the response to one question influences

the participant’s responses to subsequent questions.

This bias cannot be avoided but can be minimised by asking general questions before specific ones, positive

questions before negative ones and behaviour-related questions before attitude-related questions.

Biased reporting 

Occurs when some findings of the study are not equally

represented in the research report.

Reexivity. Also independent researchers may be asked to review the results (researcher triangulation).