Resourcd File

720 views

Published on

0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
720
On SlideShare
0
From Embeds
0
Number of Embeds
22
Actions
Shares
0
Downloads
0
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

Resourcd File

  1. 1. Research Methods PSYA4
  2. 2. The major features of science • Empiricism – Information gained through direct observation or experiment. • Objectivity – Observations and experiments should be unaffected by bias (such as researcher expectations). • Replicability – It is important that research can be repeated and similar results obtained, this adds to the reliability of the study. • Hypothesis testing – This means using the scientific process to test a hypothesis under controlled conditions. • Theory construction – Observations help to construct theories to understand the phenomena around us.
  3. 3. The scientific process Induction Testable hypothesis Deduction Observations Testable hypothesisConduct a study to test the hypothesis Conduct a study to test the hypothesisDraw conclusions Draw conclusionsPropose theory Propose theory Observations
  4. 4. Theory construction Induction • Involves reasoning from the particular to the general. • For example a scientist may observe instances of a natural phenomenon and come up with a general law or theory. • Before the twentieth century, science largely used the principles of induction- making discoveries about the world through accurate observations, and formulating theories based on the regularities observed. • Newton’s Laws are an example of this. He observed the behaviour of physical objects and produced laws that made sense of what he observed. Deduction • Involves reasoning from the general to the particular. • Starts with a theory and looking for instances that confirm this. • Darwin’s theory of evolution is an example of this. He formulated a theory and set out to test its propositions by observing animals in nature. He specifically sought to collect data to prove his theory.
  5. 5. Can psychology claim to be a science? • Some psychologists are subjective (this means using their own opinion when conducting research). These methods aim to be valid, but data collected using interviews, questionnaires, content analysis or observations which are then triangulated could be less objective than experiments. • The scientific method is reductionist and deterministic. This is because we are attempting to explain complex human behaviour in simplistic terms. • Psychological treatments of mental health problems have had modest success; therefore the goals of science are not always appropriate in psychology. • Experimenter bias and demand characteristics make studying human behaviour less objective and this compromises validity.
  6. 6. On the down side .. Do we really want to be a science? • The scientific approach is reductionist, simplifying complex phenomena and theories down to basics. • Science is also determinist in its search for causal relationships, i.e. if X determines Y. • Science also takes the nomothetic approach - looking to make generalisations about people and find similarities. • Some psychologists argue the idiographic (individual) approach is more suitable when treating patients. Currently psychology has only moderate success when treating mental illness. • Qualitative research is seen as less than scientific but triangulation can make this method more objective and valid.
  7. 7. Validating new knowledge using peer review • Peer review is the assessment of research by others who are experts in the same field (peers). This is usually done before research is published. • This is an essential check to prevent incorrect or faulty data from entering the public domain. • It is also necessary where any application for funding is concerned so it affects not just the researcher but also the university department that employs them. public bodies allocate funding to the most worthwhile research. • Every researcher should be prepared for their work to be scrutinised. Peer review is a way of establishing the validity of scientific research. • There are online journals (e.g. ArXiv) where readers rate the articles and can form the basis of peer review.
  8. 8. Advantages of peer review • It ensures that any research conducted and published is of high quality • Checks the validity of the research • Judges the credibility of the research and assessing the quality and appropriateness of the design and methodology. • Peers assess how original the work is and whether it refers to relevant research by other psychologists. • Make a recommendation as to whether the research paper should be published in its original form, rejected or revised in some way. • Helps to ensure that any research paper published in a well-respected journal has integrity and can, therefore, be taken seriously by fellow researchers and by lay people.
  9. 9. Criticisms of peer review • Unachievable ideal – it isn’t always possible to find a suitable expert to review the report. • Anonymity – research is conducted in an environment surrounded by competition. Relationships between experts sometimes affect objectivity. Now journals prefer open reviewing. • Publication bias – peer review tends to prefer positive results. • Preserving the status quo – peer review tends to prefer results that support previous theory, rather than going against it. Science does not really like huge shifts in opinion or theory. Peer review is... 1. Slow 2. Expensive 3. Profligate of academic time 4. Highly subjective 5. Prone to bias 6. Easily abused 7. Poor at detecting defects 8. Useless at detecting fraud
  10. 10. Practice Questions 1. What is science? (2) 2. Explain two of the key features of science? (3+3) 3. Outline the three goals of science? (3) 4. Outline the scientific process (hypothetico-deductive method) (4) 5. Explain the principle of falsifiability (2) 6. Explain the 3 stages in the development of a scientific discipline/paradigm? (2+2+2) 7. List 2 criticisms of the scientific approach (3+3) 8. Why may the scientific nature of research be reduced because it is being carried out on human beings? (3+3+3) 9. What is meant by the term peer review in validating new scientific knowledge? (3) 10. Name the sections of the report to be sent for peer review?(2) 11. Outline 3 problems with peer review (2+2+2) 12. Choose two psychological approaches and explain, with examples, why they are scientific (3+3) 13. Choose one psychological approach and explain, with examples, why it is not scientific (3) 14. Give 2 points to support and two to criticise Psychology as a science (2+2+2+2)
  11. 11. Layout of a Psychological Investigation 1. Abstract – A summary of the study, covering 2-5. 2. Introduction/aim – What the researchers intend to investigate, refer to previous studies, and state the hypotheses. 3. Method – What the researchers did, in enough detail to be replicated, also mentioned are the participants, the environment, the procedure, the data collection methods and the instructions to the participants. 4. Results – Statistical data and descriptive statistics. 5. Discussion – An explanation of the results is given, with implications for future research. 6. References – Full details of the journals mentioned.
  12. 12. Designing Psychological Investigations Consent form Standardised instructions Standardised debrief Instructions should be clear and succinct. They must explain the procedures of this study relevant to participants and include a check of understanding of instructions. They should also use language appropriate and be as courteous as possible. This is not a consent form so explicit references to ethical considerations are not necessary It is perfectly acceptable to include comments such as 'you are free to withdraw from the study at any time.' This is where all is explained to the participant . They are thanked at the end of his/her contribution. The aim of your study is included and also a reference to ethical guidelines. Refer to participant’s results being confidential. Make sure that participants’ names are not showed/written on the results sheets, raw data or anywhere else in the report (re: confidentiality). Display the standardised instructions to the participant(s) again. The form should contain some of the following: •The purpose of the study •The length of time required of the participants •Right to withdraw •Reassurance about protection from harm e.g. the availability of medical supervision •The requirement to undertake a series of psychological tests •Reassurance about confidentiality of the data Ethical and methodological issues need to be included for participants to make an informed decision
  13. 13. Reliability • Inter-observer reliability – When there are two or more observers you need to ensure there are general agreements in the observations. 0.80 or more is good. • Internal reliability – The measure of something that is consistent within itself (e.g. the questions on an IQ test measure IQ). • External reliability – The measure is consistent over several occasions (e.g. the same individual gets the same IQ score). Methods used to assess/ensuring reliability • Inter-rater reliability – Two or more interviewers/observers must get the same outcome on 80% or more of the behaviours. • Split-half method – Compare an individual’s performance on two halves of a test. • Test-retest method – A person repeats a test a month or so after doing the test the first time.
  14. 14. Validity • Internal validity – Whether the researcher measured what they intended to measure. – It is affected by extraneous variables. Changes in the DV can be due to EVs rather than the IV. • External validity – The extent the results of the study can be generalised to others (also known as ecological validity). Methods used to assess/ensuring reliability • Face validity – Does the test look as though it measures what it intends to measure. • Concurrent vali/dity – Results from a new test can be compared to a previously well-established test.
  15. 15. Sampling •Use the people available at the time. But can be biased. Opportunity •Selected by advertisements. Biased due to the high motivation. Volunteer •Participants are identified and then selected by a lottery method or a random name generator. Normally unbiased.Random •Sub-groups are identified (e.g. boys or girls), then a predetermined number from each group is selected randomly in proportion to the target population. Stratified • Same as above but with opportunity sampling.Quota • Participants recommend friends to join the study. Prone to bias because you only get access to a small proportion of the population. Snowball
  16. 16. Ethical Issues with Human Ppts • Same as AS (informed consent, right to withdraw, debrief, deception, protection from harm etc) • What you need to know now is the BPS code of conduct: 1. Respect: the researcher needs to maintain privacy, confidentiality, and have informed consent. You would not need informed consent in an observation where people would normally expect to be observed. Deception is only acceptable when revealing the reasons for the study affect the integrity of the results. 2. Competence: psychologists should maintain high standards. 3. Responsibility: psychologists have responsibilities to the clients, the public, and the science of psychology, this includes a debrief and protection from harm. 4. Integrity: psychologists should be honest and accurate, including the reporting of findings. Any misconduct should be reported to the BPS.
  17. 17. Ethical Issues with Non-Human Animals Why would you choose to study using non-human animals? Because animals are fascinating Animals allow for greater control and more objectivity You can use animals when you can’t use humans (e.g. Harlow’s monkeys) There is enough of a similarity between humans and animals to draw conclusions from one to the other
  18. 18. Ethical Issues with Non-Human Animals Can you justify using animals in research? Sentient beings Do animals feel pain? There is evidence that they do…but is this the same as being conscious? There is evidence that primates have self- awareness Speciesism Singer (1990) said that testing on different animals is no different to sexism or racism. However Gray (1991) says that we have a duty to humans. Animal rights Singer’s view is utalitarian (whatever is best for the greater good is acceptable), so if an animal reduces human suffering it is justifiable. Regan (1984) says animals should never be used under any circumstances. Animal research is strictly controlled and you need a license. Licenses are only granted if: 1. Results are important enough. 2. Research cannot be done without animals. 3. There is a minimum number of animals used. 4. Discomfort is kept to a minimum
  19. 19. Ethical Issues with Non-Human Animals • Russell and Birch (1959) proposed the 3Rs –Reduction –Replacement –Refinement • The House of Lords endorses this principle. • But still in the UK we use animal research, as the law states that drugs need to be tested on at least two species of live mammal.
  20. 20. Graphical representations Top Tips for graphs Give the graph a title. Label both of the axes. Use a ruler. Make the graph at least half a side in size. Make sure you plan the graph before rushing in. Think carefully about your choice of graph (don’t be tricked like you were before ).
  21. 21. Key Terms for Statistical Analysis Probability • Psychologists look at data to see if the pattern of results could have occurred by chance. If there results did not occur by chance then we say they are significant. Significance • You need to have a null hypothesis (H0) and an alternative hypothesis (H1). What we are looking for is a significant (large) difference in results so that the differences seen in our samples are different and not due to chance; we want to accept the alternative hypothesis. Chance • Normally psychologists set the probability level a p≤0.05 which means there is a 5% possibility the results occur by chance in the sample, when there was no real difference in the results in the general population. Observed value • The rho or u value calculated is called the observed value. Critical value • You need to look in a table of critical values to see if the results are significant. You need to know the 1) degrees of freedom (df) – normally the number of ppts in a study (N); 2) one- or two-tailed test; 3) significance level – normally p≤0.05
  22. 22. Type I and II errors • Type I accepting the alternative hypothesis when you should have rejected it and accepted the null hypothesis. » Type I errors are common when the significance level is too high (e.g. 10%) • Type II accepting the null hypothesis when you should have rejected it and accepted the alternative hypothesis. » Type II errors are common if the significance level is too low (e.g. 1%)
  23. 23. Choosing statistical tests You use chi-square... ...if the data have been put into categories, they are classified as nominal data. This is the test of a difference or of an association. The results are independent in each cell, and the expected frequencies (i.e. number of entries) in each cell are greater than 4. You use Spearman’s rho... ...if a test of correlation is needed as the hypothesis predicted a correlation. The data involved ratings made by participants that are ordinal/interval data. Each participant has two co-variables. You use Mann–Whitney... ...if a test of difference is required because the hypothesis predicts there will be a difference between the two groups. The design is independent groups, and the data were scores on a test (ordinal or interval data). You use Wilcoxon... ...if a test of difference is required because the hypothesis predicts there will be a difference between the two conditions. The design is repeated measures as all participants were tested twice. Interval or ordinal data has been collected.
  24. 24. R • Some tests are significant if the observed value is greater than the critical value, while some tests are the reverse. • In the chi-squared and Spearman’s rho the obtained (observed) value has to exceed the critical value for an effect to be significant. • In the Mann-Whitney and Wilcoxon tests the observed value has to be below the critical value for a significant result. • You will also find this under each table. • But try to remember this: – If there is an R then the observed value should be gReateR than the critical value (e.g. Spearman’s and chi- square). If there is no R (e.g. Mann-Whitney and Wilcoxon) then the observed value should be less than the critical value.
  25. 25. Wilcoxon T • H1 = Participants recall significantly fewer emotionally threatening words than neutral words (1-tailed) • H0 = There is no significant difference in the number of emotionally threatening words and neutral words recalled • N = 9 N = the number of participants • T = 7 • Critical value = 8 • p0.05 (significance level) • Is it significant?
  26. 26. Chi Square • H1= There is a significant association between subject studied and personality (2- tailed) • H0 = There is no significant association between subject studied and personality • χ2=3.22 • df= 1 (degrees of freedom) (Row – 1) x (columns – 1) • Critical value = 3.84 • p0.05 (significance level) • Is it significant?
  27. 27. Mann Whitney U • H1= Children using the maths scheme attain significantly higher scores than children not using the maths scheme (1-tailed) • H0 = There is no significant difference in scores between children using the maths scheme and children not using the maths scheme • N1 = 9 (number of people in the smaller group) • N2 = 10 (people in larger group) • U = 8 • Critical value = 24 • p0.05 (significance level) • Is it significant?
  28. 28. Spearman’s Rho • H1 = There is a significant positive correlation between test scores and the amount of time spent studying for the test (1-tailed) • H0 = There is no correlation between test scores and the amount of time spent studying for the test • N = 10 (number of pairs/ppts) • r = 0.88 • Critical value = 0.564 • p0.05 (significance level) • Is it significant?
  29. 29. Qualitative Data • Qualitative researchers dislike quantitative data because the results are not applicable to everyday life. • Qualitative methods more subjective as the results aim to represent the real world. • The qualitative researcher collects subjective data using broad questions that allow the respondent to answer in their own words. They may also observe the behaviour indirectly (through looking at other’s notes) or directly. • Data sets are very large (but may be through small samples). • Qualitative data should not be converted to numbers.
  30. 30. EXAMPLE OF QUALITATIVE DATA ANALYSIS – CONTENT ANALYSIS
  31. 31. Step 1 – Coding • This is the process of identifying categories, themes, phrases, or keywords in the data set. – If a psychologist was carrying out an observation he may identify a number of categories and then allocate an individual observer to analyse each one. – If a psychologist had conducted an interview, the researcher identifies a theme (e.g. being upset) and then analyses the entire script to find more examples of this theme. • Coding is a thoughtful process and not at all superficial. The categories or themes are decided upon in two ways: – Top-down approach (thematic analysis) – Bottom-up approach (grounded theory)
  32. 32. Step 2 – Analysing the data • Top-down approach (thematic analysis) – When you analyse the data you find themes and concepts by using an existing theory/explanation. – For example, the clinical characteristics of schizophrenia may be used as categories when coding self- descriptions of patients with schizophrenia. • Bottom-up approach (grounded theory) – When analysing the data the codes and categories emerge from the data. So codes remain grounded in the observations rather than being generated before the study. – You would use grounded theory in an area of psychology which is new or when developing new insights.
  33. 33. Step 3 – Summarising the Data • This is a process where it makes it possible to turn qualitative data into quantitative data. • Behavioural categories that are identified are listed and used later when summarising the data. • For examples, when analysing data the psychologist lists the themes or categories, or gives examples of behaviour within each category using quotes from participants or descriptions of typical behaviour in that category. The psychologist then counts the frequency of each occurrence, turning qualitative into quantitative data. The psychologist can then draw conclusions.

×