Practical Issues in Research

Question 1

Practical Issues

Answer 1

• Considered after the Research Question is decided and method planned out
• Practical issues focus on method like design decisions, variables being measured, sampling to find participants and the materials and procedures required to gather the data

Question 2

Design Decisions

Answer 2

Dependent on Research Question:

• If exploratory and not much is known about area = Case Study (Produces in-depth qual data that can be analysed)
• It focuses on the description of an area= Ob or Q
• Explanation of Area of Interest= Experiment (Cause and effect conclusions, statistical significance)
• Improving something=Treatment considered with a control group

Question 3

Variables

Answer 3

• How to operationalise variables as needs to be measurable
• What variables need to be measured and what variables need to be controlled (c and e)
• Exploratory study tends to look for themes that will reveal variables of interest

Question 4

Sampling

Answer 4

• Depends on the study as it may need to be specialised (eg women and aggression over memory which is universal)
• Random least biased and most rep but not easy as people need to be available and know about it to take part
• Opportunity is straightforward (Phobias won’t be found like that unless sought after)

Question 5

Materials

Answer 5

• Materials needed to gather data and deciding how materials will be gathered
• Recorder, pen, paper, video, letters
• Qs need to be prepared, interview schedules, suitable story (Data collection tools)
• Avoid Bias. DC needs to be considered so can be avoided without compromising the study

Question 6

Collection and Analysis

Answer 6

• Observations (what type of observation)

- What sort of analysis (inferential test)

Question 7

Social Psychology Practical Issues

Answer 7

• Both P and O are abstract concepts so how they are measured and whether the measure is valid need to be considered
• Lab Experiments
• Questionnaires
• Sherif

Question 8

Social Psychology: Lab Expirements Practial Issues

Answer 8

• Volunteers tend to share characteristics so makes it hard to generalise to the wider pop (sampling) Milgram used an AD though attracted a wide range all were white and from the same area and willing to participate
• The use of Standardised Procedure means all practical issues are considered in advance and it can be ensured that the study is replicated. Milgram replicated as did Burger though changed aspects.
• Labs are artificial environments so all variables can be controlled to establish C and E about something affecting P or O but this can be costly and long (Variables and Funding) Milgram conducted at Yale.
• Does not reflect real life behaviour as highly controlled and aware of the control. Low EV. DC’s impact the conclusion of whether something effects O or P (Milgram did not tell his participants until after to avoid this but told immediately after and eased distress)
• Task Validity. O operationalised into volts administered in Milgram
• Will cause psychological harm when investigating O and P. Ethically unsound. Will need money to do follow up like Milgram did

Question 9

Social Psychology: Questionnaires

Answer 9

• Low Response Rates is a high practical issue as well as everyone who does respond to having the same traits. Limiting the ability to apply the results to the wider pop
• Quant data produced is quick to analyse and compare. So not a big practical issue
• Replication on a mass scale to test for consistency. Can be costly and faces the same low response rates
• Self-report natures lead to biases which are a practical issue. Social desirability and acquiesce bias
• Could include an open question which could lead to practical issues as it takes longer to analyse and is open to interpretation and consequently subjectivity

Question 10

Social Psychology: Burger

Answer 10

-Replication of Milgram
-Burger used the same verbal prods and the same generator, as well as the same idea of the ‘victim’, is a confederate.
-Made changes to make more ethical as it was a modern replication. There was a two-step screening process to filter out anyone who might be unduly stressed by the experience. The participants were warned 3 times in writing that they could withdraw at any point and still keep the $50
-Burger is trying to make the study more ethical by not putting participants through (what he regards as) unnecessary distress. However, his assumption that participants who would go to 165V would go all the way to 450V is a big assumption.
-Concludes that Milgram still stands
Burger found that 70% of participants in the baseline condition were prepared to go past 150V, compared to 82.5% in Milgram’s Variation #5. Not statistically significant given the number of people involved. But it is based on an assumption
-Lacks EV/Reliable/Bigger sample with a wider age and included women

Question 11

Social Classic Study: Shreif

Answer 11

-22 boys is not a large sample. Any anomalies skew the results. However, Sherif went to lengths to screen, removing any from troubled backgrounds (Costly so high PI)
Only boys were used, so the results may not generalise the girls or. All children, so the results may not generalise to adults. (High PI as not gen) The boys were “all American” types: white, bright and sporty. Not representative of America
- Low PI as used standardised procedure that could be replicated such as the bean-counting test along with the tournament and the prizes. However, other procedures were developed by Sherif “on the fly” as events developed (the boys themselves requested the baseball match and Sherif had to intervene to prevent a fight). These things might happen differently if the study was replicated again. Also high as had multiple researchers and had to use multiple data collection methods. (Number score and recordings)
-Independent Variables were the stage of the experiment. (formation, friction and integration). Dependent variable measured by observing the boys behaviour and friendship patterns and tape recording their conversations and recording the phrases they used; also the boys filled out questionnaires on their attitudes to their own group and the other group. Subjective. No c and e.
-No DCs as in natural environment so low PI. Has ecological validity, because these were real boys at a real summer camp, doing real activities. Even the specially created tasks (fixing the broken water pipe) seemed real. Some unrealistic features, such as the camp counsellors not intervening until the boys were actually ready to fight.
-High as boys were matched on IQ, sport and background in order to remove partipant variables but added extra time and limited sampling method
-Only two weeks so not real life prejudice
-No C and E as limited control

Question 12

Practical Issues in Cognitive Psychology

Answer 12

The main topic Memory is an abstract concept it cannot be seen so it is hard to measure and study.

• Lab Experiments
• Case Studies
• Baddeley (1966)

Question 13

Cog Psychology: Lab Experiments

Answer 13

• Low PI as memory is assumed to be universal
• Standardised Procedures mean low PI as easy to replicate. (Seb-Hern data can be compared with results on a previous intelligence test and English data and similar patterns in development have been found - that digit span increased to age 17. So credible)
• SP means EVs are also accounted for (Seb had strict control like no learning difficulties or year repeats ). So C and E could be established
• SP and strict controls saved cost and time. (570 volunteers aged 5-17 years in part one. One sequence per second, if right another added, started with 3 sequences of 3 digits)
• DCs and not real life memory as conducted in an unnatural setting and unnatural memory test (Bartlett War of the Ghost also conducted in a lab but the use of folk tales, because they are written in an unusual style, may not actually represent everyday memory but is more valid than studies that have used nonsense material.)

Question 14

Cog Psychology: Case Studies

Answer 14

-High PI as not generalisable. Small group or one person who isn’t representative by nature. (HM damaged hippocampus in surgery. Left unable to make new memories. But still had a lot of memories from before his surgery, which suggests he still possessed LTM, but could no longer add to it.) Unusual case. Not rep of wider pops memory
-Low PI. Brain scans are easy to reproduce and objective. Part of SP
-High PI as can’t replicate due to it being unethical to do so and to cause the same memory impairments through things like Brain damage
-Produces vast qual data. Takes time to analyse and could be done subjectively
-Low task validity as memory lists do not reflect real life memory. (HM showed improvements on the performance of new
skills such as reverse mirror-drawing in which he had to acquire new eye-hand coordination)

Question 15

Cog Psychology: Baddeley

Answer 15

• Low PI as a large sample of 72 so anomalies cancelled out/But conditions broke down (15 did AS condition)/ All Brits and volunteer sample might have more people with particularly good memories who enjoy doing memory tests which is not representative of people in general.
• SP means low PI. Same words every three seconds. Test for consistency.
• Low PI as easy to analyse and compare. Displayed on charts that showed. Semantically similar words do seem to be confusing and the experimental group lags behind the Control group. In fact, the experimental group never catches up with the Control group and performs worse overall than the Acoustically Similar group
• Helps with a revision to know STM encodes acoustically and LTM semantically. Make semantic not acoustic links
• High PI as conducted in controlled conditions in order to establish C and E which costing time and money
• High PI.the ecological validity of this study is not good. Recalling lists of words is quite artificial but you sometimes have to do it (a shopping list). Recalling the order of words is completely artificial and doesn’t resemble anything you would use memory to do in the real world. Surprise 5th is more like real life.

Question 16

Practical Issues in Learning Psychology

Answer 16

• Main topics include Aggression and Phobias which are abstract concepts that can’t be seen and are hard to measure and study effectively
• Lab Experiments
• Animal Research
• Observations
• Watson and Rayner

Question 17

Learning Psychology: Lab Experiments

Answer 17

• High PI as volunteers share the same characteristics so it is harder to apply to the rest of the population as they don’t rep them (Capafons used 41 people with a fear of flying)
• SP means it can be replicated easily to test for consistency (Bandura’s procedure is very reliable because it can be replicated – as Bandura did, replicating the study in ’63 and ‘65. This was easy to do because of the standardised procedure, same script, same checklist categories, etc).
• High PI as control over variables, establish c and e over something affecting learning but costs time and money.
• High PI. EV issues as an artificial environment that is unnatural to ppt and so learning behaviour of aggression or phobias will not reflect real life. (Bandura lack validity. Children were put in a strange situation, exposed to some unusual adult behaviour and given toys to play with which encouraged them to act unnaturally. A Bobo Doll is designed to be hit and knocked over children would suppose the experimenters wanted them to play with the Bobo Doll in this way. DCs)

Question 18

Learning Psychology: Animal Research

Answer 18

• High PI. Qual differences between them. Humans are more complex. Despite our evolutionary similarities, we cannot generalise from animal experiments to humans because our thought processes and behaviours are different. If this is true then all animal studies are “low-quality research” and fail Bateson’s decision-making cube test.
• High PI. Gaining the right amount of animals and the right species being difficult means smaller samples due to Home Office restrictions. Research needs a licence from the Home Office; the premises must be licensed for animal research as must every individual involved in the research. Laboratory animals must be procured from “high-quality suppliers” who comply with Home Office standards.
• Low PI. Animals having faster breeding cycles and so replication is easier and quicker. Their faster breeding cycles make it possible to observe development in an experimental timeframe of days or weeks, rather than the years it would take for humans.
• Low PI. Due to animals being less likely to show DCs as they will be unable to guess the aims of the research. There are practical advantages to such research. Animals can be controlled more exactly than humans and observed more continuously. Their lack of self-awareness reduces the likelihood of demand characteristics in experimental conditions.
• Open to Researcher bias. Interpreting animals as can’t communicate. High PI.It is also argued that animals have different needs and perceptions from humans, so we cannot know to what extent they are suffering. This makes the Bateson decision-making cube and attempts to “minimise” harm impossible to carry out.

Question 19

Learning Psychology: Observations

Answer 19

• High PI. Share same characterises as opportunity samples are just there at the same time so similar traits. Or those who are helpful
• High PI. Sample size in covert observation when ppts are debriefed and may drop out
• Structured followed procedures like recording data when something specific happens or at a certain time (Bandura)
• EV is high as a familiar setting. So behaviour reflects real life aggression or fear
• No control in naturalistic environments. Limits C and E ability so cannot conclusively say something is linked to aggression
• High practical issues due to covert observations being unethical. A deception so needs to spend time in the debrief counteracting this.

Question 20

Learning Psychology: Watson and Rayner

Answer 20

• High PI. Only a single ppt of the 9-month-old baby from America which cannot be generalised to other countries, genders and or ages. Unrep but chosen for his normalcy and fearlessness
• Low PI. SP used and it was carefully documented (right down to the numbers of days and the time of day) and it was filmed. For ethical reasons, the study hasn’t been replicated, but it could be replicated quite easily. So no practical issues
• Inter-Rater reduces bias reducing PI problems. The surviving film of the procedures means we can all view Albert’s responses and see his fear for ourselves. So objective.
• Helped developed therapies like Flooding based on CC to remove phobias through association
• High PI due to no control group. No C and E between loud noise causing the phobia. It might have been due to repeated exposure. But tested before so it wasn’t there previously.
• High PI with EV due to taking place in setting that is unfamiliar. The setting for the experiment lacks ecological validity because Albert was away from his playroom and familiar nurses. This may have made him nervous.
• High PI. Unethical as it caused Albert a lot of stress and breaking many ethical guidelines that would stop the study being replicated. They didn’t extinguish his fear reaction, possibly leaving Albert with long term phobias. This is ignoring the principle of reducing harm.

Question 21

Practical Issues in Biological Psychology

Answer 21

• Aggression is an abstract concept as can’t be seen and so it is hard to measure and study
• Scanning Techniques
• Twin and Adoption Studies
• Correlational Research
• Raine

Question 22

Brain Scanning (CAT, PET and fMRI)

Answer 22

• High practical issues due to being expensive and so only a small sample will be gained (CAT scans involve having a person lying in a huge scanner and the scanner moves around them, sometimes dye is used for 30 minutes, costing up to £600)
• Low practical issues as the images it produces allow multiple researchers to analyse and come to the same conclusion so can be compared easily (PET scanning is R because it can be repeated and the same results can be found. Whenever someone speaks the same area of the brain is used. The same areas of the brain are consistently found for different activities, so R)
• High Practical issues due to being in a lab-based environment that has high control over all variables allowing to establish cause and effect between something and brain activity which would cost time and money
• Not ecological. So High PI. Doesn’t reflect real life so their brain activity won’t reflect to real life activity. (Raine: The level of violence was not controlled. NGRI’s could have murdered somebody with poison, a gun, by strangulation or with a chainsaw. All of these various methods of dispatching a victim involve different degrees of violence.)
• High PI. With task. (continuous performance task of spotting a target) to access brain activity. Reaction and not aggression.

Question 23

Twin and Adoption Studies

Answer 23

• High PI. Twins have very different upbringings to the rest of the population who have only one sibiling/none/ older siblings. Twins may have unusual lives in other ways too. They attract attention and often get treated the same way even if they have different personalities; they may be mistaken for one another. This may cause them to start behaving more like each other than other brothers or sisters do.
• Low PI. Scientific methods like DNA sequencing which provides objective quant data which is easier to analyse and compare to judge whether DZ or MZ.
• High PI. MZ pair being exposed to different environmental triggers for one gene so it is harder to compare. Just because twins share the same home and parents, it doesn’t mean their upbringing has been exactly the same. They may have different friends, interests and relationships and these differences may grow more pronounced as the twins grow older. The psychological terms for this are that the twins may share the same genotype (biological similarity) but they may not share the same phenotype (social similarity).
• High PI. Adoption studies have no control over EVs. Like why they were adopted or upbringing so it is harder to compare. There are lots of confounding variables in adoption studies. Adoption agencies try to match children to adoptive families as similar as possible to their biological family (in terms of race, ethnicity, class, etc) and this makes it harder to tell if upbringing is at work.
• Low Practical issues due to no research bias in measuring/operationalisation of variables as twins are naturally occurring. Twins are an example of a naturally-occurring variable being perfectly changed for study: MZ twins share 100% of their genes, DZ twins share about 50%. Because researchers are not manipulating this variable themselves, it reduces the risk of researcher bias.

Question 24

Correlational Research

Answer 24

• Low PI. Use Quant data with the expression of correlations on a graph that can be compared and analysed quickly and easily. Which can lead to new research as They are carried out with relative ease and input from a researcher (for example, nothing needs to be set up artificially as they use pre-existing data). Correlations can, therefore, can decide if new research should be generated to investigate relationships that may not have been thought previously.
• High PI. They cannot infer cause and effect. It is impossible to claim that one co-variable actually causes the other co-variable as it could be a third unknown variable (a mediating variable) that causes both variables to change together. This may lead to erroneous conclusions and misunderstanding. For example, there may be a correlation between daycare and children’s behaviour, yet we cannot be sure that daycare leads to behavioural problems in children, or whether intervening variables play a part (e.g. mothers that send their children to daycare may be more stressed which may have an impact upon behaviour)
• High PI. Self-report methods lead to various biases. Like SD and A.

Question 25

Biological Psychology: Raine

Answer 25

• High PI. NGRIs are unusual offenders. They are people who have killed someone, but either don’t remember doing it or are too confused to stand trial. These people are not representative of “typical” murderers, still less of typical violent individuals. As Raine points out, not all of the NGRIs killed their victims violently. Also non-murderers so won’t represent murderers.
• High and Low. Used PET scans. PET is a reliable brain imaging technique that has been used (with growing success) since the 1970s. It produces objective and replicable results and it can be tested and re-tested to check its reliability. But expensive. The results were sometimes unclear and had to be interpreted, which introduces subjectivity and low reliability.
• Not fully accountable for actions (NGRI murderers) due to brain differences.
• Low PI. Stopped taking medication to allow for normal functioning, allowing for c and e.
• High PI. The CPT used by Raine could be criticised for being artificial and unconnected to violence or provocation. The participants were all doing an unusual task and in an unusual state of mind when the PET test was carried out. This lowers the ecological validity of the study. 32 minutes of an irrelevant task that had no violence so did not trigger brain areas associated with it.

Question 26

Practical Issues in Criminal Psychology

Answer 26

• Reasons for Criminality and Memory is an abstract concept that can’t be seen, measure or study
• Lab Experiments into EWT and Juries
• Field Experiments into EWT
• Loftus and Palmer

Question 27

Lab Experiments into EWT and Juries

Answer 27

• High PI. Due to mainly being volunteers for labs which wouldn’t represent a group of witnesses who happen to be there at the time.
• Low PI. SP can easily be replicated to test for consistency
• High PI due to it being hard to measure stress and attention to crime
• High PI. Low Ecological Validity. Due to the crime being staged and in controlled in condition. which doesn’t reflect the environment of the crime so it won’t replicate the real stress and attention will differ.
• High PI. A video of a crime which doesn’t reflect the realism of one taking place in the surroundings that the person is in themselves
• High PI. Replication of juries won’t feel the same pressures.

Question 28

Field Experiments

Answer 28

• Low PI. Opp samples are more rep of actual witnesses of crime who are just people who happen to be there at the time
• High PI. Less control over EVs so can’t replicate to test for consistency
• High PI. Hard to stage crime outside the lab as variables cannot be easily controlled
• Low PI. Witnesses natural environment so natural stress and attention
• Low PI. If unaware then natural stress and attention that will reflect real life as DCs are avoided.

Question 29

Criminal Psychology: Loftus and Palmer

Answer 29

• High PI. Used 45 American students who were the same age so not rep. Students are not representative of the general population in a number of ways. Importantly they may be less experienced drivers and therefore less confident in their ability to estimate speeds. This may have influenced them to be more swayed by the verb in the question.
• Low PI. Easy to replicate (i.e. copy). This is because the method was a laboratory experiment which followed a standardised procedure. 7 films of traffic accidents, ranging in duration from 5 to 30 seconds, were presented in a random order to each group.
• Low PI. Gathering Quant (Estimate speed of the car). Easy and quick to analyse. The participants in the “smashed” condition reported the highest speed estimate (40.8 mph)
• Useful. Informs the police that leading Qs effect recall so during interviewing this should be avoided.
• High PI. Lacked mundane realism / ecological validity. Participants viewed video clips rather than being present at a real-life accident. As the video clip does not have the same emotional impact as witnessing a real-life accident the participants would be less likely to pay attention and less motivated to be accurate in their judgements.
• Not real life stress and attention so don’t reflect real life.

Question 30

Clinical Psychology Practical Issues

Answer 30

• Mental Health is an abstract concept that is hard to define, measure and study
• Case Studies
• Longitudinal Studies
• Cross-Cultural Studies
• Rosehan

Question 31

Case Studies

Answer 31

• Unique Person with unique symptoms of MI and situations so cannot represent everyone with that disorder. 25% of ppts with Szh don’t respond to drugs. High PI.
• Low PI. Some of the same methods are used that use quant data allowing for easy comparison (Psychocmetrict testing and the GAF scale in the DSM when diagnosing)
• High PI. The complex and unique combination of methods to gather data means it is hard to compare ppts. triangulation.
• High PI. Gathers in-depth data about symptoms and situations of the patient. Unstructured interviews result in qual data that is open to subjectivity and bias when interpreted.
• High PI. Unethical as ppts are vulnerable and may feel obligated to partake even if they feel distressed during the evaluation.

Question 32

Longitudinal Studies

Answer 32

• High PI. High drop out rates so smaller cohort harder to compare
• High PI. Complex so needs to be adaptable which means that the replicating conditions for consistency are hard.
• Low PI. Cohort effect. Changing factors as the study progress making shared identities different over time
• High PI. Qual data. in-depth detail into circumstances of MI and it’s progression over time which means it will be longer to analyse which will incur more costs.
• High PI. Unethical invasion of privacy. Vulnerable patients may want to drop out. Should the researcher get involved to help the participant with their problem (which ruins the reliability and validity of the study and perhaps goes against guidelines for professional boundaries between practitioner psychologists and their clients)

Question 33

Cross-Cultural Studies

Answer 33

• Low PI. Due to representing cultures and how MI may vary as a result.
• High PI. Small samples don’t represent everyone.
• High PI. Misommination between countries due to linguistic differences may affect the consistency of results as each country do it differently. So data is inconsistent.
• High. No SP. As trying to accommodate differing cultures. Diff researcher as women wants a female.
• Low PI. Encourages mindfulness of researcher to different cultures which can cause subjective interpretation on behaviour/Still an element of bias as it is hard not to compare to your own culture and what you see as normal.

Question 34

Rosenhan

Answer 34

• Low. Rosenhan made a point of using a range of psychiatric hospitals - private and state-run, old and new, well-funded and under-funded - from across the United States. So rep.
• High. Nevertheless, 12 is a small sample for a country as big as the USA and a few “bad apples” could have skewed the results of Rosenhan’s observations. Ethnocentric and doesn’t represent any other country or any time period outside the 1970s
• Low. SP used as all patients described the same symptom of hearing ‘thud’, as well as recording data for a structeed observation of thing, like how many times the nurses interacted. Meaning the study could be accurately replicated.
• Low. Gathered Quant data from how many times the nurses and psychiatrists interacted with patents which can easily be analysed and compared. Psychiatrists stopped and talked 4% of the time while nurses only 0.5% of the time
• Useful as highlighted how unfair and poor the conditions inside these hospitals were. Over-medication instead of finding ways to treat them and improvement of diagnostic practises.
• Low PI. High EV. Staff were assessed in a familiar environment and were unaware of the study so their behaviour would be free of DCs and reflect real life. Only Rosenhan informed hospital Administrators
• High PI. Lawyers had to be on hand to remove any records of them having a MI and to get them out if needed. Which would have been expensive as they charge by the hour. Could not have been done without funding.