Paper 2: Research Methods (Year 2)

Question 1

What’s content analysis?

Answer 1

Content analysis is a way of analysing qualitative data.
This could include studying a conversation, speech, email, letter, newspaper film of a book.
The aim of content analysis is to summarise this information so that conclusions can be drawn.

Question 2

How do we convert quaulatative data to quantative data?

Answer 2

Coding is the initial stage of content analysis. Where the research needs to decide on pre-determined categories.

Some information may be very large such as interview transcripts, and need to be categorised into meaningful units.

This may involve counting the amount of times a particular word or phrase appears in the text. For example, newspaper reports may be analysed to see the number of times negative words are used to refer to people with a mental illness.
The researcher would then count the amount of times words such as “mad”, “nuts” or “crazy” appear in the text.

This allows the researcher to convert qualitative data to quantitative data

Question 3

How does thematic analysis help with content analysis?

Answer 3

Content analysis may also involve generating qualitative data.

A theme refers to any idea that keeps appearing in a bit of text or speech. These are likely to be more descriptive than just one word and may include phrases like “the mentally ill are a threat to the wellbeing of our children”.

Such themes may be developed into broader categories such as “stereotyping of the mentally ill” and “treatment of the mentally ill”.

Once enough themes have been identified to cover most aspects of the data being analysed, the researcher may collect a new set of data to test the validity of these categories and themes or make an overall conclusion.

Question 4

evaluation of conent analysis

Answer 4

No ethical issues - A strength of content analysis is that there are no ethical issues associated with using it as a research method. This is because a lot of the material that a researcher wants to analyse already exists in the public domain.
Researcher’s do not need to get consent to analyse the information.

Multipurposeful
Another strength is that content analysis is flexible. This is because it produces both quantitative and qualitative data depending on the aim of the research.

aa weakness - lack of objectivity
People are studied indirectly as part of content analysis and the information produced is studied outside of the original context.
The researcher may attribute opinions to the speaker that are inaccurate.
This means that the researchers own biases may influence the outcome of the content analysis, making the data less valid and more subjective as one researchers view may differ from another.

Question 5

wha is reliability

Answer 5

Reliability refers to how consistent the findings from an investigation or measuring device are.
A measuring device is said to be reliable if it produces consistent results every time it is used. If the results can be repeated; it is reliable.

Question 6

Ways of assessing reliability

Answer 6

THE TEST RETEST METHOD
This involves giving the same test or questionnaire to the same people on different occasions.
If the questionnaire or test is reliable, then it should produce the same (or similar) results each time it is used.
This method is mainly used with questionnaires and psychological tests (such as IQ tests) but can also be used with interviews.
In the case of a questionnaire or test the two scores taken at different times would be correlated to see if they are similar. If they are similar, then it is reliable.

Question 7

WHATS INTER OBSERVER RELIABILITY

Answer 7

To make sure that an observation is reliable, inter-observer reliability must be established.

This involves having two or more observers record the same event but recording their data independently.

The data collected by the 2 observers should be correlated to see if it is similar. This can also be done with a content analysis (inter-rater reliability) and interviews (inter-interviewer reliability).

Question 8

Validity

Answer 8

Validity refers to whether a test, observation or experiment produces a result that is true or genuine.

Question 9

knternal validity

Answer 9

This refers to whether the effects observed in an experiment are due to the manipulation of the independent variable and not some other factor.
If Demand characteristics are present in a study, this could reduce the internal validity of the research.
Did the experimenter measure what they intended too

Question 10

extrnal validity

Answer 10

This relates to generalising the results to other settings, people (population validity) and time periods.
Ecological validity is all about generalising the results from one setting to other settings.
Natural experiments are not always high in ecological validity just because they take place in a natural setting. This is because the task used to measure the dependent variable may not be a task we use in everyday life (low mundane realism) and this can lower ecological validity.
For example, a research may five people at the bus stop a list of words to learn. The setting is natural as it is a field experiment, however, the task is not. This lowers ecological validity.

Question 11

population validity

Answer 11

Population validity – Looks at whether, results from one sample of P’s applies to other people.

Question 12

temporal validty

Answer 12

Temporal validity

This refers to whether the findings from a study or the concept from a theory are applicable over different time periods.
For example, the Asch study is seen to be outdated as it took place in the 1950’s when people were more conformist that they are now.

Question 13

face validty

Answer 13

This is the least sophisticated measure of validity. Face validity is simply whether the test appears (at face value) to measure what it claims to.
This can be determined by merely looking the instrument or passing it to an expert to check.
For example, if an IQ test asked “what is your favourite chocolate” then this would lack face validity.

Question 14

INCOMPLETE IMPROVING VALIDTY

Answer 14

QUESTIONNAIRES
INTERVIEWS
EXPERIMENTS
OBSERVATIONS

Question 15

Concurrent validity

Answer 15

Concurrent validity is a measure of how well a particular test correlates with a previously validated test.
A new IQ test may be given to a group of participants then scores may be correlated with the same participants IQ scores on a previous, well established IQ test.
Close agreement between the two sets of scores would show high concurrent validity.
This would need to be a positive correlation of +0.80

Question 16

WYAS OF IMPROVING VALIIDITY - 7 WAYS

Answer 16

Control Group
In experimental research, a control group can be used to improve the validity of an experiment.
Using a control group means that the researcher is able to assess whether any changes in the dependent variable were actually due to the independent variable.

Standardised procedures
Standardised procedures (same instructions, same researchers) can also be used to minimise the impact of investigator effects on the outcomes of a study.

Single blind procedures
The use of single blind procedures means that participants are unaware they are taking part in a control group or experimental condition until the end of the study. The researchers are the only ones that know which condition the participants are in.
Information is kept from the participants to reduce any bias in the results.

Double blind procedures
Double blind procedures are also used to improve the validity of a study. This means that both the participants AND the researcher are unaware of which participants are the control group or the experimental group.
This reduces participant bias and investigator effects

OBSERVATIONS - In order to get high ecological validity in observations, Covert observations are used as these ensure that the behaviour observed is authentic and genuine as they are unaware of their role in a study.
In addition, good behavioural categories will improve the validity of the information gathered in an observation.

QUALITATIVE METHODS- Because qualitative methods include a lot of detail, they are assumed to have higher validity than quantitative methods of research. For example, a case study is a true reflection of the participants reality compared to simply getting that information from a questionnaire.
Interpretive validity also needs to be high. Interpretive validity is the extent to which the researcher’s interpretation of events matches the participants.
This can be demonstrated through things such as including direct quotes from the participants in research.

TRIANGULATION - Triangulation also improves validity. This means that data is used from a number of sources not just one. For example, interviews with friends, personal diaries and observations

Question 17

THE SCIENTIFIC METHOD

Answer 17

The scientific method has 3 parts to it.

1. Observation & description of a phenomenon or group of phenomenon
2. Formulation of a hypothesis and use of a hypothesis to make predictions
3. Performance of experimental tests (+ form conclusions).

Science therefore involves making predictions, tested by scientific observations under controlled conditions.
In this way, theories and hypotheses can be validated (found to be true) or falsified (found to be untrue).

Question 18

REPLICABILITY

Answer 18

This involves repeating research to check the validity of the results.
If a study can be repeated over a number of different contexts and circumstances, then we can see the extent to which findings can be generalised.
In order for replicability to become possible, research must be fully and clearly written up so that it can be repeated under controlled conditions.
If you can get the same results twice it implies the results are valid
It is unlikely that someone will get the same results twice due to chance alone.
If something cannot be repeated it is thought to indicate flaws.

Question 19

Objectivity & the empirical method

Answer 19

Objectivity means that a scientific researcher must not allow personal opinions or biases to influence the data they collect or influence the behaviour of the participants that they are studying.
To lessen the possibility of bias, researchers use standardised instructions, double blind techniques etc (refer back to previous notes).
Objectivity is the basis of the empirical method. The empirical method or empiricism emphasises the importance of data collection based on direct sensory experience and not influenced by any bias.
The experimental method and the observational method are examples of the empirical method in psychology.

Question 20

FALSIFICATION

Answer 20

Part of the validation process of psychology is falsification or falsifiability.
This means that a scientific theory or hypothesis must be empirically testable to see if it is false. Replication is the accepted way of determining this.
Popper stated that even when a theory or hypothesis has been successfully and repeatedly tested, it is not necessarily true, it has just not been shown to be false!
Theories that survive the most attempts to falsify them are the strongest. Not because they are necessarily true, but because they have not yet been proven false.
This is why psychologists often say “this supports” or “this suggests” as opposed to “this proves”.
For example, many critics point out that Freud’s theories are not supported by any experimental data and state that his ideas cannot be supported by any real evidence. In order for a theory to be scientifically valid, it must be possible falsify it with experimental evidence, and many of Freud’s notions are not falsifiable.

Question 21

Theory construction and hypothesis testing

Answer 21

A theory is a set of general laws or principles that can explain particular events or behaviours.
Theory construction occurs through gathering evidence via direct observation for example, having a “feeling” or “guessing” that STM has a limited capacity based on the observation that people can’t remember when bombarded with lots of information.
It should also be possible to make predictions on the basis of a theory.
This is the role of hypothesis testing. Hypothesis testing involves testing a theory to see if it will be supported or refuted. If it is supported, the theory will be strengthened. If it is refuted, the theory may need to be revisited.
Deduction is the name given to the process of deriving new hypotheses from an existing theory.
Theory/prediction – Men and women can be just friends. How would you test this?

Question 22

paradigm shifts

Answer 22

A paradigm is a set of shared assumptions and methods.
An important change that happens when the usual way of thinking about or doing something is replaced by a new and different way

The philosopher Thomas Kuhn suggested that, unlike other science subjects like biology or physics, psychology does not have 1 paradigm. In psychology, there are many different theories and assumptions about behaviour; not just one.
Because of this, Kuhn stated that psychology should be classed as a “pre-science”.
According to Kuhn, progress within a scientific subject happens when there is a scientific revolution.
This can happen when a small group of researchers begin to question the accepted paradigm which eventually results in a paradigm shift.

Question 23

LENGTHY QUESTIONS

Reporting Psychological investigations

Answer 23

abstract - This is a summary of the research usually consisting of information about the previous research done on that particular topic, the aims, hypothesis, methodology, results, conclusions of the current research and suggestions for further research.

Introduction
This section includes a literature review (background research) of the theories, concepts and studies related to the subject of the report.
This section starts off with broad theories gradually becoming more specific.
Then the aims and hypotheses of the research in question are presented.
method
Design – information about the experimental design is included (e.g. repeated measures) as well as information about the type of experiment (observation, field etc.). Reasons for choosing the design are also given here.
Sample – The number of participants, sampling method, target population and biographical and demographic information about the sample is given here.
Apparatus/materials – description of any equipment involved and how it was used.
Procedure – This includes a step by step breakdown of what happened in the study. This section also includes everything that was said to participants in the briefing, the standardised instructions given and the debriefing.
Ethics – included in this section is an explanation of how any ethical issues were addressed in the study.
This section summarises the findings of an investigation. Included in this section are descriptive statistics such as tables, graphs, measures of central tendency and measures of dispersion. Inferential statistics should also be included. This includes reference to the statistical test that was used, critical values and calculated values, levels of significance and which hypothesis was accepted and rejected.
If the researcher used qualitative methods, this section would include the thematic analysis/content analysis

discussion
This has 4 sub-sections
1) Explanation of the findings – a summary of the findings in verbal form
2) Relationship to background research – a discussion of what was found in relation to previous research findings
3) Limitations and modifications – possible errors such as poor sampling, lack of control over variables etc. are discussed as well as possible ways to rectify these faults in the future
4) Implications and suggestions for future research – Further studies that could be undertaken are presented here as well as other ways of testing the hypothesis. Also, any implications and applications that the findings of the present study suggest are presented here.

Conclusion
A paragraph summarising key conclusions drawn from the study.

References
A list of all websites, books and articles is included here. Tips for how to reference can be found using this website www.neilstoolbox.com . To do a reference follow this format –
SURNAME, First letter of their first name. (Year of publication). Title of article. Publisher’s name.
E.g. Knowles, B. (2009). Single ladies. Elektra records.

Appendices
This section contains everything else that was used in the study; Instructions given to the participants, raw data, calculations of the statistical tests plus any other stimulus material used.

Question 24

correlaion studies

Answer 24

Correlation co-efficient
This is a number between +1 and -1.
The closer the number is to 1 (either +/-) the stronger the correlation.
The closer to 0 the weaker the correlation.
The +/- only indicates the direction of the relationship (negative or positive correlation).

Question 25

what is a stastical test?

Answer 25

A statistical test is used to determine whether a difference or an association found in a particular investigation is statistically significant (as opposed to have occurring due to chance).
Once you have determined whether a test is significant you can find out whether you accept or reject the null hypothesis.

Question 26

how to choose a stastical test

Answer 26

1. Determine whether a researcher is looking for a difference or correlation.
2. What experimental design was being used?
   Independent group design (aka unrelated design)
   Repeated measures design (aka related design)
   Matched pairs design (aka related design)
   (Please note this stage does not apply to correlations as there is no design for such studies).
3. Determine the level of measurement.
   Nominal
   Ordinal
   Interval

Question 27

designs

Answer 27

Unrelated design – Refers to cases where participants from each experimental group are different, hence unrelated.
Related design - Refers to instances where the same participants are used in all conditions or people who are closely matched on certain variables.

Question 28

what are the thre type of measurement or data?

Answer 28

nominaldata
ordinal data
interval data

Question 29

what is nominal data

Answer 29

This is data that is represented in the form of categories.
For example you can count how many males and females work in a store. How many people said yes or no? How many people picked red/blue/purple/green as their favourite colour.
In nominal data each participants response can only appear in one of the categories e.g. a participant would either have said ‘yes’ or ‘no’.
use the ‘mode’

Question 30

what is ordinal data

Answer 30

This data is usually ranked (put in order) and usually involves a rating scale. 
For example if a class was asked how much they like psychology on a scaled of 1 to 10 one being a little and 10 being a lot. The data does not have equal intervals e.g. someone who rates ‘8’ on the psychology scale does not mean they like the course twice as much as someone who rated it as ‘4’. 
Ordinal data is also subjective (unsafe measure) as it is based on opinion for instance what constitutes as ‘4’ or ‘8’ on that psychology scale could be different for each person.
use 'median' & 'range'

Question 31

what is interval data

Answer 31

Interval data
This data is based on numerical scales that includes units of equal and precisely defined sizes (unlike ordinal data).
This data will usually involve things you take measurements with e.g. a stopwatch, thermometer or weighing scales. For example if a researcher recorded how long it took for participants to complete a written recall test in psychology.
It is a precise and sophisticated form of data (safe measure) and is necessary for parametric tests (test which are more able to determine if there is significance within a data set).
use ‘mean’ & standard deviation

Question 32

what is the choosing a statistical test mnemoic.

Answer 32

FOR NOMINAL DATA
Chatting (CHI-SQUARED // used for INDEPENDENT GROUPS or UNRELATED DESIGN)
Shit (SIGN TEST // used for REPEATED MEASURES or REPEATED DESIGN)
Can (CHI-SQUARED // used to TEST FOR A CORRELATION)

FOR ORDINAL DATA
Make (MANN-WHITNEY // used for INDEPENDENT GROUPS or UNRELATED DESIGN)
Women (WILCOXON // used for REPEATED MEASURES or REPEATED DESIGN)
Seem (SPEARMAN’S RHO // used to TEST FOR A CORRELATION)

INTERVAL DATA
Unpredictable, (UNRELATED T-TEST// used for INDEPENDENT GROUPS or UNRELATED DESIGN)
Ruthless, (RELATED T-TEST// used for REPEATED MEASURES or REPEATED DESIGN)
Paranoid (PEARSON’S R // used to TEST FOR A CORRELATION)

Question 33

what is statistical tests

Answer 33

Used in Psychology to determine whether a significant difference or correlation exists (which will determine whether the null hypothesis is rejected or accepted).

Question 34

whats the diff betwee te experimental hypthhesis and the null hypothesis

Answer 34

Experimental/alternative hypothesis
Researchers begin their research by writing an experimental/alternative hypothesis. This may be directional (if they feel confident about likely outcome) or non-directional (if they are uncertain what the outcome will be).
You should only choose a directional hypothesis if past research suggests a likely outcome in the scenario.
The null hypothesis
This states there is ‘no difference’ between the conditions (or ‘no correlation’ between two variables if the study was looking for a relationship between two variables).
The statistical tests determines which hypothesis is accepted and which is rejected. If significant results – Accept experimental hypothesis and reject null. If not significant, accept the null + reject experimental

Question 35

what is the pobability and significance of statisticl tests

Answer 35

Statistical tests work on the basis of probability rather than certainty. All statistical tests employ a significance level – the point at which the researcher can claim to have discovered a significant difference or correlation within the data (this will also help determine which hypothesis to reject and accept).
The usual level of significance in psychology is 0.05 (or 5%).
This is written as p _< 0.05
This means that if a researcher finds a statistical difference/ correlation there is up to a 5% probability that the results were actually due to chance (it was a fluke).

Question 36

lower level of significance

Answer 36

metimes a more rigid (strict) level of significance may be used e.g. 0.01 (instead of 0.05).
Using a 0.01 level of significance means that if there is a difference/correlation you can be more certain that the result is statistically significant, and there is only a 1% probability that the results were due to chance.
This level of significance is used in studies where there may be a human cost e.g. in drug trials.

Question 37

WHATS THE CALCLATED VALUE

Answer 37

The calculated value/observed value
In your psychology exam the stats formula for the calculated value would have been completed for you.
To check for statistical significance, the calculated value they provide you with must be compared with a critical value. This will tell you which hypothesis we accept and reject.
Each statistical test has its own version of table of critical values. You will be expected to work out the critical value in the exam (for this you need to refer to the caption under the critical value table which will be provided in the exam).

Question 38

what. are the 3 criterias for working out the critical value

Answer 38

To work out the critical value there are 3 criteria:

1. One tailed or two tailed? If the hypothesis was directional this means that your hypothesis was one-tailed, if there was a non-directional hypothesis then that means you used a two-tailed hypothesis.
2. The number of participants in the study. This usually appears as the N value on the table. For some tests degrees of freedom (df) are calculated.
3. The level of significance (probability value) is p _<0.05

Question 39

what are type 1 errorss

Answer 39

his is referred to as an optimistic error – because you wrongly accepted the experimental hypothesis (alternative) hypothesis as true and wrongly rejected the null hypothesis.
However the null hypothesis was actually true. In these cases the researcher claims that they have found a significant difference/correlation when one does not exist.
We are more likely to make a type 1 error if the level of significance is too lenient (too high) e.g. 0.1 or 0.10 instead of 0.05.

Question 40

what are type 2 errors

Answer 40

This is referred to as the pessimistic error – this is because the researcher wrongly accepts the null hypothesis and rejects the experimental hypothesis. Therefore they are claiming there was no significant difference/correlation when actually there was.

We are more likely to make a type 2 error if the level of significance is too strict (too low) e.g. 0.01 (as potential significant values may have been missed).

Psychologists favour the 5% (0.05) level of significance as it best balances the risk of making a type 1 or type 2 error

Question 41

when would you use a mann-whitney test

Answer 41

When to use a Mann-Whitney test
If you are looking for a difference between two groups, the level of data was ordinal and the design was unrelated (independent group design).

Question 42

wilcoxon

Answer 42

This is used when testing for a difference, a related design (repeated measures design) and the level of data is ordinal