Research methods part 2

Primary data

Data gained directly from participants, usually from participants doing questionnaire, interview or observation

Strength of primary data

Researcher has control of data so it can be designed to fit aims and hypothesis of study

Weakness of primary data

Requires time and effort, and is expensive

Secondary data

Data that’s been collected by someone other than the person conducting the study, usually from journals, books, or websites

Strength of secondary data

• Cheap and easily accessible

• Data has probably already been statistically tested and peer reviewed

Limitations of secondary data

• Content of data may not exactly fit with needs of study - may be incomplete or out dated

• May be variation in quality and accuracy of data

Meta analysis

Analysing data from a large number of studies which have involved the same research questions and methods

Define peer review

• The process of subjecting a piece of research to independent scrutiny

• By other psychologists working in a similar field who consider the research in terms of its validity, significance and originality

Main aim of peer review

• Allocation of research funding

• Assess the quality and relevance of the research

• Suggesting improvements - suggest minor changes to work to improve report that’s been submitted

• Assessing the research rating of university departments - funding unis get depends of reviews of the peer review process

Evaluation of peer review

• Finding an expert - not always possible to find an appropriate expert to review research proposal

• Anonymity - although good to maintain objectivity and honesty of appraisal, reviewers can use their anonymity to criticise rival researchers

• Publication bias - truthful research could be ignored as editors of journals usually want to publish significant results

• Burying ground breaking research - peer review process may suppress ground breaking research that may contradict views of researcher

3 measures of central tendency

Mean, median, mode

Strength of mean as a measure of central tendency

• Most sensitive as it includes all values so more representative of set of scores

Weakness of mean as a measure of central tendency

Easily distorted by extreme values

Strengths of using the median

• Not effected by extreme scores

• Once arranges in order the median is easy to calculate

Weakness of using the median

Not as sensitive as mean as not all values are included in final calculation

Strength of using the mode

• Very easy to calculate

• Unaffected by extrme values

Weakness of using the mode

Not very useful if there are several modes

2 measures of dispersion

Range and standard deviation

When and why is the range useful?

• Useful when assessing how representative the median is

• The higher the range the less representative the median value is as it would indicate scores are spread widely from that figure

Standard deviation

Tells us how far scores deviate from the mean

Interpreting standard deviation

• High standard deviation = greater spread of scores around the mean, meaning not all participants were affected by the IV in the same way

• Low standard deviation = scores are more clustered close to the mean, meaning most participants behaved in similar way

Strength of using the range

Very easy to calculate

Weakness of range

Only considers 2 most extreme values which may be unrepresentative of the whole data set

Strength of standard deviation

More precise measure of dispersion as it includes all values in final calculation

Weakness of standard deviation

• Easily distorted by a single extreme value

• Can be complicated to carry out by hand

Describe normal distribution

Mode median and mean are the same value

Describe positive skew

• Mode, then median, then mean

• Frequency decreases from mode to mean

Describe negative skew

• Mean, then median, then mode

• Frequency increases from mean to mode

Assessing reliability of observations

Calculate inter observer reliability

Improving observational reliability

• Multiple observers

• Operationalised variables

• Train observers

• Do pilot study to know if observers are applying categories properly

Assessing reliability of self report techniques

• Split half method - split the test in half and have participants do both halves, if results are the same for both halves it means it has internal reliability

• Test-retest - external reliability - give participants same test on 2 occasions after a week or 2 has passed so they forget answers. If results are the same then it has external reliability

Improving reliability of self reports

• Make questions clear and precise

• Pilot a questionnaire to check if questions are clear enough

• Use same interviewer with each participant or fully trained if using more

Assessing reliability of experiments

Carefully controlled procedures, instructions and conditions

Improving reliability of experiments

• Use exactly the same procedure for all participants

• Use same conditions

• Replicate in exactly the same way if replicated by other researchers

Assessing validity

• Face validity - does the test look like its measuring the dependent variable

• Concurrent validity - Compare your test with well established one and get participants to do both. If they get similar scores then ur test has concurrent validity

Improving validity of questionnaires

• Assure responses are anonymous

• Review questionnaires or tests when they have low face or concurrent validity

Improving validity of experiments

• Use control group so researcher can clearly see changes in DV is due to IV

• Standardise procedures to reduce investigator effects

• Reduce demand characteristics by using double blind / single blind research

Improving validity of observations

• High ecologically validity if covert

• Make sure behavior categories are not too broad, overlapping or ambiguous

Type 1 error

• Falsely rejecting null hypothesis

• False positive

Type 2 error

• False negative

• Falsely accepting null hypothesis

Content in an abstract

• 150-200 words

• Quick picture of study and its results

• Very briefly mention aims, hypothesis, method, results & conclusion

Contents in an introduction

• Review of previous research

• Begin broadly then narrow down to hypothesis

• State aim, research prediction and hypothesis

Contents of method

• Detailed description of what researcher did

• enough detail so it can be easily and precisely replicated

• Include: Design, participants, materials used, procedure, ethics

Contents of results

• Summarise key findings from investigation

• Include descriptive stats - include central tendency and dispersion, tables, graphs

• Inferential stats with justification

Discussion

• Summary of results - results and what they tell us

• Relationship to background research - whether the study is in line with what research in the intro suggests

• Limitations and modifications - strengths snd weaknesses of research and how it could be improved

• Implications and suggestions for further research

Referencing journals

Authors name, date, title of article, journal title, volume (issue number), page number

Referencing books

Authors name, date, title of book, place of publication, publisher

7 features of science

• Theory construction

• Hypothesis testing

• Empirical method

• Falsifiable

• Objectivity

• Replicability

• Kuhn - paradigm shifts

How does a paradigm shift occur

• Normal science → model shift - evidence against pre existing paradigm starts

• Scientists add on extra bits to theory, believe new research is flawed, don’t believe theory has been falsified

• Model drift → model crisis → model revolution

• Mounting evidence against paradigm & new paradigm explains things previous paradigm couldn’t

• Model revolution Mounting evidence→ paradigm change