Research Methods (MET) Flashcards

Question

Identify what quasi- experiments CAN establish about group differences, and what they CANNOT establish

Answer 1

That there are group differences * ...but not why

Answer 2

Lab * Easy ... * ...as pps are recruited for study Field * Unlikely... * ...as pps are usually unaware they are in experiment Natural * Depends... * ... on whether pps are recruited or if data is collected without their knowledge

Answer 3

* Involves making ppts aware of any aspect of a study that would affect their willingness to participate * e.g., the aims, its procedures, their rights and what their data will be used for * Ppts can then make an informed judgement about whether they would like to take part. * Ppts should be issued with a consent letter ... * ...detailing all relevant information that might affect their decision to participate. * This is then signed by the participant (if old enough to understand) and parent (if the ppt. is under 16)

Answer 4

* Asking other people who could have taken part in the research... * ... to decide whether the deception / stress is justified * Cannot be presumed to be representative of the population * Not actually going to take part, so less concerned about ethics of study

Answer 5

* Getting permission to deceive a pp * Usually consent is obtained as part of being involved in a number of experiments, e.g., as part of a course * Pps may be wary knowing there is a possibility that each experiment is 'the one with the deception / stress' * Affecting behaviour in all experiments

Answer 6

Obtain consent after the study for use of data * Not really consent... * ...just right to withdraw data

Answer 7

* Deliberately misleading pps during study... * ...or being misleading by withholding information * Acceptable only if there is a 'strong scientific or medical justification' for deception * In which case, at the end of the study, give ppts. a full debrief to make them aware of the true aims.* * This debrief includes asking for retrospective consent. / giving right to withdraw data

Answer 8

* Participants should not be placed at risk of physical or psychological harm * Study should be no more stressful than everyday life * If participants have been subject to stress, the researcher should provide counselling, if they require it * Debriefing is not an acceptable way of 'dealing with' protection

Answer 9

* Pps have right to control information about them * Others should not know (or be able to work out) identity / behaviour of pps * Anonymity actively maintained in databases * Researchers usually do not publish names

Answer 10

* Pps have the right to cease to participate in a study... *...at any point and without giving a reason * This should be explained at the outset * Pps should also have option to withdraw data at end

Answer 11

* Acceptable if they could normally expect to be observed in the situation * ...i.e., it is in a public place * e.g., in (some) covert observations / field experiments

Answer 12

British Psychological Society (BPS)

Answer 13

* Independent groups * Different pps in each condition * Repeated measures * Pps take part in all conditions * Matched pairs * Pps matched with someone similar in other condition(s) on a relevant participant variable / individual difference

Answer 14

Individual differences * Pps differ in many ways ... * ...which may interfere with effect of IV on DV Order effects * Events in experiment occur in an order... * ...so earlier events may affect later events

Answer 15

* Strength: no order effects * Limitation: individual differences between pps * Control: random allocation to conditions

Answer 16

* Strength: no individual differences between pps * Limitation: order effects possible * Control 1: counterbalancing * 1/2 of pps take part in conditions in one order, 1/2 the other way round * Control 2: randomisation * Items from conditions are randomly mixed together (e.g., in Stroop task)

Answer 17

* Avoids order effects (unlike repeated measures) and * Minimises individual differences (unlike independent groups) but * Expensive and time- consuming to use

Answer 18

* Each pp in the sample could be given a number... * ... which is put into a hat / computer program. * The first 50 numbers drawn out of the hat / by the software ... * ...would be allocated to one of the groups.

Answer 19

* The sample would be randomly allocated by hat method/software... * ... to take part in conditions in one of two orders * Condition A then condition B or * Condition B then condition A * Order effects occur in both conditions / directions.... *... but cancel each other out

Answer 20

* Mix items together in a single sequence... * ... rather than pps taking part in one condition in its entirety then the other condition. * Order of items random and unpredictable/ pps switch between conditions continuously * Order effects occur randomly in both directions.... *... but cancel each other out

Answer 21

* It is rare for conditions to be divisible into small units... * A condition usually consists of one 'big' thing that happens

Answer 22

* Matching: put all pps in order on a specified variable * e.g., age, aggression. * Pair each pp with their closest counterpart. * Randomly allocate from within pairs to conditions

Answer 23

* All members of the target population have an equal chance of being selected * Obtain a sampling population, i.e., a list of possible pps to select the sample from (Ideally, this is the population, in most cases, not!) * Give everyone in sampling population a number * Use a random number generator to select N pps

Answer 24

Stratified sampling * The sample reflects the proportions of people in certain sub groups (strata) within the target population * Population is divided into strata groups that the researcher wants represented * Then random sampling is used to choose pps within each strata

Answer 25

Systematic sampling * A quasi- random technique * Every nth member of the target population is selected * e.g. every third house on a street

Answer 26

Opportunity sampling * The researcher uses anyone who is willing and available at the time to take part * e.g., academic uses her students

Answer 27

Volunteer sampling * A volunteer sample involves participants selecting themselves to be part of the sample * e.g., by responding to an advertisement

Answer 28

* Population = group that the research is about, i.e. to whom the study's findings should apply * Sample = group who take part in a research investigation. * Sample is taken from a target population * It is presumed to be representative of that population.

Answer 29

* Generalisation = extent to which the findings and conclusions can be applied to the population. * Sample representative of population.=> generalisable * Bias = when certain groups are over or under represented within the sample selected * Bias => lack of generalisability

Answer 30

* Random sampling is free from researcher bias... *... as the choice of pps is left to chance rather than actively done by the researcher * So, random sample => more likely to be representative

Answer 31

* Convenience * Using whoever is available is much easier than ... *... using special procedures to choose pps

Answer 32

* Stratified sample more likely to be representative / generalisable * The effect of random error is reduced by stratification * Groups represented in terms of proportionality in population

Answer 33

* Volunteers may not be representative of population, * e.g. they might be more helpful

Answer 34

More practical than random sampling... * ...especially when it doesn't really matter if sampling is not done entirely randomly * ...e.g., names from school rolls

Answer 35

Independent measure Repeated measure Match pairs Quasi-experimental design

Answer 36

Using different participants file each condition of the experiment. The advantage is that there are no order effects

Answer 37

An experimental design where the same participants take part in each condition of the experiment. Advantage- control of participant variables and fewer participants needed so less expensive. Disadvantage- order effect and demand characteristics.

Answer 38

An experimental design where pairs of participants are matched in terms of key variables such as age or IQ. Advantage- experimental control to reduce error Disadvantage- there may not be a suitable match

Answer 39

* Statistic that identifies the strength ... * ...and direction of a linear relationship between two co-variables * Plotted on a scattergram

Answer 40

* A correlation does not involve manipulation * There is no IV and DV but there are co-variables

Answer 41

a continuous variable that is not part of the main experimental manipulation but has an effect on the dependent variable. such as personality

Answer 42

* You cannot be sure about cause and effect * Variable A may have caused B to change but you cannot be sure * It is possible that B caused A to change or that C caused both A and B to change

Answer 43

* Ethical reasons * Can do a correlation when it is not acceptable to manipulate variables... * ...because it would not protect pps from harm * Practical reasons * Not possible to manipulate an IV / do an experiment * e.g., if pps would not follow instructions in a field experiment

Answer 44

* Questionnaires * Interviews

Answer 45

Open Closed

Answer 46

Structured Semi- structured Unstructured

Answer 47

* They are less time- consuming than other methods * Data on large number of participants can be gathered quickly compared to (e.g.) experiments * People are often rich, valid sources of information about their own lives. * This would be impossible to measure without asking them.

Answer 48

* Self- reports are not actual behaviour and ... * ...what people say they do is not necessarily what they (would) do (in real life or experiments) * Pps may not know / remember the truth.. * ...or lie / display social desirability effects

Answer 49

* Do not have a fixed range of answers * Respondents are free to answer in any way they wish * Produce qualitative data that is rich in depth and detail

Answer 50

* Offer a fixed number of responses * Can be yes/no or scale of 1-10 * Produce numerical data which is easy to analyse

Answer 51

* They provide depth / detail and ... * ...greater diversity of responses / more meaningful information in the response * They avoid participant frustration associated with fixed choice responses. * Participants can report in their own words rather than be 'hemmed in' by the closed set of responses.

Answer 52

* They yield numerical data that can be analysed statistically. * This allows researchers to test hypotheses * They should be more reliable and comparable between participants. *Closed questions can give precise meanings to each response

Answer 53

Quantitative * Numerical data... * ...scored / counted / measured for each participant Qualitative * Non-numerical data... * ...that is expressed in words

Answer 54

Test hypotheses.

Answer 55

.. interpretation

Answer 56

* Usually have more detail than quantitative data... * ...which is a number rather than a deep exploration. * More likely to measure what it is supposed to measure, i.e., have greater validity

Answer 57

* Questionnaires can obtain detailed data by simply being longer... * ...and the number can be a more valid indicator than ambiguous words.

Answer 58

a measure of how test performance predicts behaviors in real-world settings

Answer 59

a type of external validity that refers to the validity of the findings in relation to the progression of time. For example, do the findings of conformity research (e.g. Asch, 1951) still apply today?

Answer 60

whether you can reasonably generalise the findings from your sample to a larger group of people

Answer 61

the degree of confidence that the causal relationship being tested is trustworthy and not influenced by other factors or variables.

Answer 62

refers to the extent to which results from a study can be applied (generalised) to other situations, groups or events.