Research Methods- all content

Research Methods

• Experimental method: types of experiment, laboratory and field experiments; natural and quasi-experiments and their strengths and limitations

  • Laboratory experiment

    • Conducted under controlled conditions where the researcher manipulates the independent variable

    • + highly controlled

    • + establish causation

    • - demand charactersistics

    • - low ecological validity

  • Field experiment

    • Conducted in a natural environment where the researcher manipulates the independent variable

    • + high ecological validity

    • + fewer demand characteristics

    • - extraneous variables

    • - hard to replicate

  • Natural experiment

    • Conducted when the IV is naturally occurring

    • + high ecological validity

    • + fewer demand characteristics

    • - difficult to establish causation

  • Quasi experiment

    • Involve studying the effects of naturally occurring IVs

    • + same as lab/field experiments

    • - confounding variables

• Observational techniques: types of observation – naturalistic and controlled, covert and overt, participant and non-participant observation and their strengths and limitations

  • Naturalistic

    • Carried out in a natural environment

    • + high ecological validity

    • + allows study that would otherwise be unethical

    • - cannot control extraneous variables

    • - difficult to replicate

  • Controlled

    • Carried out in a lab environment

    • + easy to replicate

    • + extraneous variables eliminated

    • - demand characteristics

    • - low ecological validity

  • Covert

    • Participants do not know that they are being observed

    • + less demand characteristics

    • - more ethical issues

  • Overt

    • Participants aware that they are being observed

    • + more ethical

    • - more demand characteristics

  • Participant

    • Researcher involves themselves in the observation

    • + gain a fuller understanding

    • - may be difficult to record behaviour

  • Non-participant

    • Researcher does not involve themselves in the observation

    • + easier to record behaviour more objectively

    • - may be harder to have a full understanding of the behaviour

• Self-report techniques: questionnaires; interviews – structured and unstructured, and their strengths and limitations

  • Structured questions

    • + easy to replicate

    • - socially desirable answers

  • Unstructured questions

    • + rich qualitative data

    • - difficult to replicate, not standardised

• Correlations: analysis of the relationship between co-variables, the difference between correlations and experiments, strengths and limitations

  • There are no IVs or DVs, but the two variables being measured are co-variables

  • Plotted on a scatter gram

  • Types of correlation:

    • Positive- high scores on one variable go with high scores on the other variable

    • Negative correlation- high scores on one variable go with low scores on the other variable

    • No correlation- scores are not connected in any way

  • Evaluation:

    • + allows the relationship between two variables to be examined when a controlled experiment may not be possible due to ethical or practical reasons

    • + can be a good starting point for further research

    • - it is not possible to establish cause and effect

    • - correlations can be misused

• Case studies: strengths and limitations

  • Case studies involve the detailed study of a single individual or small group.

  • It usually tends to be from an unusual case of a certain thing

  • Case studies are generally longitudinal

  • Evaluation:

    • + detailed qualitative data, avoids reductionism

    • - many case studies have ethical issues such as lack of anonymity and psychological harm

Scientific Processes

• Aims: stating aims, the difference between aims and hypotheses

  • An aim is a general statement about the intended purpose of a study

• Hypotheses: directional and non-directional

  • A prediction about what will happen in a study

  • Precise, testable statements

  • Types:

    • Directional

      • Says the direction in which the data is predicted to be

      • Used when there is supporting evidence

    • Non-directional

      • States that there will be a difference, but not which direction it will be

      • Used when there is a lack of/ contradictory supporting data

    • Null

      • States that there will not be a difference

• Sampling: population vs. sample; techniques – random, systematic, stratified, opportunity, volunteer; implications including bias and generalisation

  • Population- group of people that the researcher wants to target

  • Sample- people that represent the larger population

  • Types of sampling

    • Opportunity

      • Participants selected by convenience

      • + convenient in terms of time and cost

      • - sample is likely to be biased

    • Volunteer

      • Participants self-select/volunteer to take part

      • + participants motivated to complete the study

      • - may be biased

    • Systematic

      • A system is created to select participants eg every 3rd person selected

      • + not biased

      • - participants may refuse to take part

    • Random

      • Participants selected at random

      • + less likely to be biased

      • - participants may refuse to take part

    • Stratified

      • Participants selected to reflect the demographics of target populations

      • + representative of target populations

      • - complex and time consuming

• Pilot studies and the aims of piloting

  • Trial run of the research study on a smaller scale

  • Pilot studies aim to:

    • Find out if aspects of the design do or don’t work

    • If parts of the design make the aims of the research obvious (demand characteristics)

    • See if timings for the tasks are appropriate

• Experimental designs: repeated measures, independent groups, matched pairs

  • Independent groups:

    • Different participants used in each condition

    • + lack of order effects

    • + lack of demand characteristics

    • - larger amounts of participants

    • - unbalanced groups

  • Matched pairs:

    • Each participant matched based on key characteristics

    • + eliminates unbalanced groups

    • + lack of demand characteristics

    • - larger amounts of characteristics

    • - time consuming and difficult

  • Repeated measures

    • Same participants used in each condition

    • + fewer participants

    • + no individual differences

    • - more order effects

    • - more demand characteristics

• Observational design: behavioural categories; event sampling; time sampling

  • Behavioural categories- examples of behaviours that have been pre-recorded

  • Event sampling:

    • The collection of data every time an event happens in an observation

    • Evaluation:

      • + useful when things happen infrequently

      • - can be hard to see everything

  • Time sampling:

    • The collection of data at pre-determined time intervals

    • Evaluations

      • + reduces number of observations

      • - could miss important behaviours

• Questionnaire construction: open and closed questions; design of interviews

  • Questionnaire

    • Set of standardised questions handed out for participants to complete

    • + easily distributed

    • + standardised

    • - socially desirable answers

  • Interviews:

    • Verbal questioning of participants usually done face to face

    • + able to explain questions to ensure understanding

    • - socially desirable answers

  • Open questions

    • + produce qualitative data

    • - can be hard to format data

  • Closed questions

    • + easier to analyse

    • - produces qualitative data

• Variables: manipulation and control – independent, dependent, extraneous, confounding; operationalisation

  • IV- characteristic that is manipulated in the study that causes the DV to change

  • DV- variable that is measured that changes throughout the experiment as a result of the IV

  • Extraneous variable- any variable other than the IV that might affect the results of the DV

  • Confounding variables- when extraneous variables are important enough to cause a change in the DV

  • Operationalisation— making sure a variable being studies is clearly defined and in a form that can be easily measured

• Control: random allocation, counterbalancing, randomisation, standardisation

  • Random allocation- method used to minimise the effect of confounding variables

  • Standardisation- ensures that all procedures and instructions are kept the same

  • Counter-balancing- attempts to balance out order effects by splitting the group and completing the condition in an AB/BA order

  • Single blind- when participants are unaware of the research aims and do not know which condition they are in

  • Double blind- when neither the observer nor the participants know the true aim of the study

• Demand characteristics and investigator effects

  • Demand characteristics- clues which help a participant guess the true aim of the study

  • Investigator effects- refers to any unwanted influence of the investigator on the dependent variable

• Ethics: British Psychological Society’s code, ethical issues in design/conduct of studies, dealing with ethical issues

  • Ethics- the potential for participants to be harmed during research

  • Psychological body- group that encourages researchers to follow guidelines and ensure participants do not get harmed

  • Types of ethical issues:

    • Protection from harm

      • Participants protected from physical and psychological harm

      • Dealt with with attempts to rectify unexpected harm

    • Privacy and confidentiality

      • Personal information should be kept private

      • Dealt with by observations happening in places expect to be observed, data and names etc kept private

    • Deception

      • Not telling participants the true aim of the study

      • Dealt with with a debrief/ reconsidering how to carry out the experiment

    • Informed consent

      • Consent from people who fully understand what is happening

      • Dealt with by gaining alternative methods of consent/debrief

• Implications of psychological research for the economy

  • If more effective treatments for mental health issues are developed, more people will be in work

  • Ineffective treatments may waste time and money

  • If treatments are effective, implementing these treatments may be costly

• Features of a science: objectivity, empirical method, replicability, falsifiability, theory construction, hypothesis testing, paradigms, paradigm shifts

  • Empirical methods:

    • These methods gain information through direct observation or experimentation rather than from unfounded beliefs or claims

    • Important as people can make claims but the only way we know anything to be true was through direct observation

  • Objectivity

    • Data is not affected by the expectations and biases of the researcher

    • Data is collected under controlled conditions

  • Falsifiability

    • Theories should be testable and there should be no possibility of them being proven false

    • Even if a theory has been repeatedly tested, it still wasn’t true or proven, it had just not yet being proved false

  • Theory construction

    • The construction of a theory occurs through gathering evidence using empirical methods

    • It is possible to make clear and precise predictions on the basis of a theory

    • The processes of deriving new hypotheses from existing theories is known as deduction

  • Replicability

    • If a theory is to be trusted, it must be shown to be repeatable across a range of different contexts and circumstances

    • Replication is also used to assess the validity of a finding

  • Paradigms

    • A paradigm is a shared set of assumptions and methods

    • It has been suggested that psychology is a pre-science as it does not have a universally accepted paradigm

  • Paradigm shifts

    • Happens when an existing paradigm is questioned by a few researchers until there is too much evidence to ignore

    • A new paradigm causes a scientific revolution

• Reporting psychological investigations: sections of a report – abstract, introduction, methods, results, discussion, referencing

  • Abstract

    • A short summary of all the major elements of a report, including the aims and hypotheses, methods/procedure, results and conclusion

    • It goes at the beginning of a report, although it is usually one of the last things that are written

  • Introduction

    • Gives details of literature that is relevant to the study taking place

    • Starts with the least relevant and progresses to the most

    • At the end of the introduction aims and hypotheses are presented

  • Method

    • Should be detailed enough to be replicable

    • Split into several sub-sections

      • Design- research methods and experimental design

      • Sample- how many participants, sampling method, target population

      • Apparatus/materials

      • Procedure- everything that happened in the investigation from the participants perspective from beginning to end

      • Ethics- how ethical issues are handled

    • Results

      • Summarises key findings, including

        • Descriptive statistics

        • Inferential statistics

        • Qualitative data

      • Raw data does not go here, it goes in the appendix

    • Discussion

      • Summary of the results in words, linked back to past research

      • Limitations of the study

      • Wider implications of the study

    • Referencing

      • Format:

        • Should I Do This Like a Pro

        • Surname, Initial, Date, Title of article/book, Place published/publisher name

• The role of peer review in the scientific process

  • Peer review is the independent assessment of a research paper by experts in the field

  • Done in order to evaluate the papers quality and sustainability for publication

Data Handling & Analysis

• Quantitative and qualitative data; distinction in collection techniques

  • Qualitative data:

    • Data that consists of words/longer answers

    • + can provide large amounts of detail

    • - can be hard to analyse/display

  • Quantitative data:

    • Numerical data

    • + easier to analyse/display

    • - can be less useful without a large amount of data being collected

• Primary and secondary data, including meta-analysis

  • Primary data

    • Data collected by a research specifically for the purpose of their study

    • + can ensure data is accurate

    • - requires planning and resources

  • Secondary data

    • Data which has already been collected by someone else

    • + inexpensive, requiring minimal effort

    • - can be less accurate/relevant

  • Meta-analysis- when a wariest of studies on a particular topic area are summarised together and their findings collated

• Descriptive statistics: central tendency (mean, median, mode), dispersion (range, standard deviation)

  • Central tendency:

    • Mean- the average of all of the data

    • Mode- most common value in a set of data

    • Median- central value in a set of data

  • Measures of dispersion:

    • Standard deviation- how far on average each score is in a set of data from the mean

    • Range- how spread out a set of data is

• Analysis and interpretation of correlation, including correlation coefficients

  • Types of correlation:

    • Positive- high scores on one variable go with high scores on the other variable

    • Negative correlation- high scores on one variable go with low scores on the other variable

    • No correlation- scores are not connected in any way

  • Correlation co-efficient

    • A number between -1 and +1, telling us the strength and type of correlation

• Levels of measurement: nominal, ordinal, interval

  • Nominal- data is caatagorised into groups, but the groups have no order

    • Examples: eye colour, male/female/other

  • Ordinal- data is ranked or ordered- we know the order, but not by how much one is more than the other

    • Examples: finishing position in a race, likely scale responses

  • Interval- data is numerical, where equal intervals between values mean equal differences

    • Examples- IQ scores, temperature

• Content analysis and coding

  • A method used to analyse qualitative data

  • The researcher must decide how to systematically sample whatever for of media it is they are analysing

    • Five types of text used:

      • Written text

      • Oral text

      • Iconic text

      • Audio-visual text

      • Hypertexts (texts found on the internet)

  • The data is then coded by creating categories (by skimming the material and making a list of the main categories)

    • The categories must be operationalised, comprehensive and mutually exclusive (not overlapping)

    • Data in each category is usually quantitative (tallies), however it may be qualitative if the researcher describes some examples

  • Evaluation:

    • + inter-rather reliability can be used

    • - observer bias/subjectivity

• Thematic analysis

  • Converts qualitative data to quantitative data

  • The first step is to transcribe the data if necessary

  • The data is then read over repeatedly

  • The themes are then identifies and re-analysed so they become clear

  • The researcher can then annotate the transcript with the themes that have been identified

  • Evaluation:

    • + tends to have high ecological validity because it is based on observations of real materials

    • - process is unscientific and open to researcher bias

Inferential Testing

• Introduction to statistical testing; the sign test – when and how to use it

  • The sign test is a method used in inferential statistics to determine whether or not an observed result is significant

  • It is a non-parametric test- there is no assumption that the data will follow a normal distribution

  • It is known as the sign test as it is based on the number of plus or minus signs present in the data after the calculations have taken place

  • When to use it:

    • If the research investigates a difference (an experiment rather than a correlation)

    • Repeated measures design

    • Nominal data (data in categories, such as exercised for 30 minutes/did no exercise)

    • Whether the hypothesis is directional or non-directional

  • How to use it:

    • State your hypothesis:

      • E.g. There will be a difference in stress levels before and after CBT.

    • Find the differences:

      • For each pair (before/after), work out if the value increased (+), decreased (−), or stayed the same (0).

    • Ignore the 0s:

      • Any cases where there is no change (0) are removed from the test.

    • Count the signs:

      • Count how many + signs and how many − signs there are.

    • Find S (the sign test score):

      • S = the smaller number of + or − signs.

    • Find the critical value:

      • Use a sign test table based on the number of non-zero scores (n) and your significance level (e.g. 0.05 for 5%).

    • Compare your S to the critical value:

      • If S is less than or equal to the critical value → the result is significant(reject the null hypothesis).

      • If S is more than the critical value → the result is not significant (fail to reject the null).

• Factors affecting choice of a statistical test, including level of measurement and experimental design

  • 3 distinct criteria:

    • Have they conducted a test of difference (e.g. a lab experiment) or a test of correlation?

    • If they have conducted a test of difference, did they use an independent measures design, repeated measuresdesign, or a matched pairs design?

      • an unrelated design refers to independent measures/groups

      • a related design refers to repeated measures and matched pairs

    • Have they collected nominal, ordinal or interval data?

  Tests of Difference		Related design
  	Unrelated design		Correlations
  Nominal data	Chi-Squared	Sign test	Chi-Squared
  Ordinal data	Mann Whitney U	Wilcoxon T	Spearman's rho
  Interval data (Parametric tests)	Unrelated t-test	Related t-test	Pearson's r