Research Methods Flashcards

Question

What are the strengths of random sampling?

Answer 1

- more representative than opportunity sampling - unbiased = confounding or extraneous variables equally divided between groups = higher internal validity

Answer 2

- can take a long time - could still be unrepresentative

Answer 3

this sample is when every 'nth' member of the target population is selected e.g. every 5th pupil on a school register

Answer 4

- avoids researcher bias - more representative that opportunity sampling

Answer 5

- participants may refuse to take part - time-consuming

Answer 6

sample is a proportional representation of the target population. the target population is broken down into sub-groups, e.g. age. you then randomly sample from those groups

Answer 7

- avoids researcher bias - representative so results can be generalised

Answer 8

- time consuming - complete representation isn't possible

Answer 9

the researcher selects pps from whoever is willing and available at the time e.g. people in the street

Answer 10

- convenient = easy, quick and cheap

Answer 11

- less representative of target population as pp's all from same location e.g. one town - researcher bias

Answer 12

participants select themselves to be put into the sample e.g. from newspaper ad

Answer 13

- easy and quick - access to people you would not normally access - people are more engaged

Answer 14

- volunteer bias as volunteers tend to be more willing and eager

Answer 15

controlled environment where researchers manipulate the IV. strict control of extraneous variables

Answer 16

- high control over confounding and extraneous variables = high internal validity = cause and effect - replication to see if valid

Answer 17

- may lack generalisation - low external validity - low mundane realism - demand characteristics

Answer 18

natural environment (participants usual environment,) researcher manipulates the IV

Answer 19

- High mundane realism - High external validity

Answer 20

- cause and effect between IV and DV is more difficult to establish - cannot replicate - ethical issues = cannot consent if don't know being studied

Answer 21

IV changes naturally = would have happened even if researcher isn't there e.g. before and after natural disaster. location could be in a lab. rutter orphan studies is an example!

Answer 22

- opportunities for research that may not otherwise be undertaken - high external validity

Answer 23

- hard to generalise findings - participants may not be randomly allocated to experimental conditions

Answer 24

IV changes due to participant variables e.g. age, gender, researcher cannot randomly alloacte pp's to groups. to due with participant themselves!

Answer 25

- often carried out under controlled conditions (same strengths as lab)

Answer 26

- participants may not be randomly allocated to experimental conditions = may be confounding variables

Answer 27

questionnaires + interviews

Answer 28

no fixed range of answers and allow pps to elaborate on their answers and give more detail. they generate qualitative data, however they are difficult to analyse

Answer 29

fixed number of responses. force pps to choose an option. easy to analyse and generate quantitative data, but lack detail

Answer 30

respondents indicate their agreement for a statement using a scale of usually five points. scale ranges from strongly agree to strongly disagree

Answer 31

gets respondents to identify a value that represents their strength of feeling about a particular topic

Answer 32

list of possibilities and respondents are required to indicate those that apply to them

Answer 33

- large sample of people - representative - easy to use - researcher doesn't need any training - fixed choice questions - easier for statistical analysis as they produce quantitative data

Answer 34

- responses may not be truthful - social desirability - response bias - pp s answer in similar way e.g. always answer yes (acquiescence bias) - closed questions with fixed responses

Answer 35

predetermined list of questions asked in a fixed order

Answer 36

list of set questions but interviewer free to ask follow up questions not on list

Answer 37

like a conversation. no set questions. general aim that a certain topic is discussed. encouraged to elaborate

Answer 38

- can replicate and test responses for reliability - reduces individual differences between interviewers

Answer 39

- may restrict information collected - people may be less happy to disclose personal info in an interview

Answer 40

- more flexibility

Answer 41

- analysis is more difficult - risk of lying - interviewer bias

Answer 42

- interview schedule = list of questions to cover which should be standardised to reduce interviewer bias - quiet room = increase likelihood participant will open up - begin with neutral questions = feel relaxed and comfortable

Answer 43

- avoid jargon - avoid emotive language and leading questions - avoid double negatives

Answer 44

natural environment where the target behaviour will normally take place e.g. at home/in a nursery

Answer 45

high ecological validity = generalised

Answer 46

- cannot replicate - lack of control over extraneous + confoudning variables

Answer 47

conditions are set by the researcher who can control key variables e.g. the strange situation/bobo doll. control of extraneous and confounding variables.

Answer 48

- higher reliability - higher internal validity - control of extraneous and confounding variables.

Answer 49

- low ecological validity = cannot generalise

Answer 50

pps know behaviour is being observed = consent

Answer 51

- doesn't break ethics

Answer 52

- demand characteristics

Answer 53

pps do not know their behaviour is being observed = no consent

Answer 54

- no demand characteristics - high internal validity

Answer 55

- low ethics

Answer 56

the observer joins in with the group

Answer 57

- increased external validity

Answer 58

- researcher bias - hard to note all behaviours - lose objectivity

Answer 59

the researcher remains separate from the group they are studying

Answer 60

- less likely to have researcher bias - more objective

Answer 61

- less insight into behaviour of the group as they are removed

Answer 62

when a target behaviour is broken up into components that are observable and measurable (operationalise them) observable, measurable and self-evident

Answer 63

a target behaviour is first established then the researcher counts the number of times this event happens

Answer 64

a target individual is first established, then the researcher records their behaviour in a fixed time frame e.g. every 60 secs

Answer 65

a small-scale trail run of the actual investigation with the aim to check procedures, material and measuring scales. so researcher can change or make modifications!

Answer 66

participants are unaware of the test being conducted but the researcher is aware. reduces demand characteristics

Answer 67

neither participants or the researcher is aware of aims of the study. free from researcher bias

Answer 68

researcher investigates an association between two variables, called co-variables

Answer 69

as one co-variable increases so does the other

Answer 70

as one co-variable increases, the other decreases

Answer 71

when there is no relationship between co-variables

Answer 72

- easy to analyse - generate quantitative data - objective and reliable - may suggest ideas for future research - quick and economical as secondary data can be used

Answer 73

- cannot demonstrate cause and effect - an intervening variable could explain why co-variables are linked - co-variables need to be operationalised - when a score is manufactured there is a chance it doesn't measure what it intends to do

Answer 74

expressed in words, could be written description of the thoughts, feelings and opinions of participants

Answer 75

- Richness of data - Greater external validity

Answer 76

- difficult to analyse - conclusions rely on subjective thoughts from researcher = researcher bias

Answer 77

data that can be counted, usually given as numbers, can be analysed and converted to graphs etc.

Answer 78

- simple to analyse - comparisons between groups made - more objective, less prone to bias

Answer 79

- less detailed

Answer 80

information obtained first-hand be a researcher specifically for the experiment. data is gathered from experiment, questionnaire, interview or observation

Answer 81

- gets the target information required

Answer 82

- requires time and effort

Answer 83

information already gathered by someone else and so pre-dates the current research project. may be in journal articles, books, or websites

Answer 84

- inexpensive and easily accessible

Answer 85

- variation in quality and accuracy = could challenge validity

Answer 86

the process of combining results from a number of studies on a particular topic. aim to provide a overall statistical conclusion

Answer 87

1. detailed, in-depth 2. one person or small group 3. longitudinal 4. rare+unique case 5. traingulation 6. used when unethical

Answer 88

- high in validity - study unusual behaviour - rich in detail generate hypotheses for future studies

Answer 89

- lack genralisability - prone to researcher bias - low in validity = rely on stories and memories from family

Answer 90

how consistent the findings are - results should be the same every time

Answer 91

1. give pp the questionnaire to answer 2. wait a period of time 3. re-administer the same questionnaire to the same pp 4. check if there is a significant positive correlation between results 5. if +0.80 then it's reliable

Answer 92

1. 2 or more observers watch the same event 2. data is recorded independently 3. compare results to see if significant positive correlation 4. if +0.80 then it's reliable

Answer 93

correlational analysis. two sets of scores are correlated and correlation coefficient should exceed +0.8

Answer 94

- test-retest - unambiguous questions - closed, fixed-choice questions

Answer 95

- same interviewer - trained interviewer - no leading questions - structured interviews

Answer 96

- operationalise behavioural categories - no overlap in categories

Answer 97

- standardised procedures

Answer 98

when a measurement in a psychological test 'on the face of it' appears to measure what it sets out to measure

Answer 99

when results obtained from a particular psychological test or scale have a positive correlation

Answer 100

- control group - standardise - single and double-blind

Answer 101

- lie scale = control social desirability - anonymity

Answer 102

- covert observation - make sure behavioural categories aren't broad or overlapping

Answer 103

- use of quotes - triangulation (use of different sources)

Answer 104

- most sensitive = includes all the values = represents all the data as a whole

Answer 105

- extremely high or low values could distort

Answer 106

- extreme scores don't affect - easy to calculate

Answer 107

- less sensitive = does not include all the scores

Answer 108

- easy to calculate - useful when data is in categories

Answer 109

- several modes not useful - crude measure = most common score might not represent data as a whole

Answer 110

- quick and easy to calculate

Answer 111

- can be distorted by a single extreme value - unrepresentative of data as a whole

Answer 112

- precise = included all values within the final calculation

Answer 113

bell-shape. most people located in middle, few people at extreme ends. mean, median and mode all at highest peak.

Answer 114

a type of frequency in which long tail is on the right and most of distribution is on the left

Answer 115

a type of frequency in which long tail is on the left and most of distribution is on the right

Answer 116

- Peer review is the process of subjecting a piece of research to independent scrutiny by other psychologists (peers) working in a similar field who consider the research in terms of its validity, significance and originality. - This happens before the research can be publicised in an academic journal. - The experts should be objective and unknown to the researcher to prevent bias.

Answer 117

1. Other psychologists who are experts in the field of research or working in a similar field of research independently check the research report before deciding whether it could be published. 2. The report is scrutinised in terms of its validity, significance and originality. 3. They also assess the appropriateness of the methods and designs used by the researcher . 4. A reviewer can accept the report as it is, accept it proviso to revisions being made, suggest the author makes revisions and re-submits the report or reject the report without the possibility of revisions or re-submission. 5. The editor makes the final decision whether to accept or reject the research report based on the reviewers’ comments / recommendations.

Answer 118

1. To ensure research that is published in an academic journal is of high quality and relevance e.g. methodology, data analysis and findings are all accurate and interpreted accurately. 2. To consider proposed research designs (before they have been carried out) in terms of the aims, planned procedures and the significance of the research in society prior to allocating funding. 3. To suggest amendments or improvements in order to improve the relevance or quality of the research or reject work that is inappropriate for publication.

Answer 119

- Peer review protects the quality of published research and preserves the reputation of psychology as a science. This encourages funding for future research proposals.

Answer 120

- Bias - Reviewers tend to be critical of research that contradicts their own view or goes against existing theory and more favourable of research that matches it. This could mean that peer review may have the effect of slowing down the rate of change within a particular scientific discipline. File-drawer problem - bias towards publishing studies with positive results, i.e., those supporting the hypothesis, but negative findings are just as important if we are to achieve a balanced view of research. Negative findings tend to be either rejected or are never submitted for publication.

Answer 121

abstract, introduction, method, results, discussion, referencing

Answer 122

- A summary of all the major elements - Identify if report if worthy of reading

Answer 123

Literature review / aims / hypothesis

Answer 124

Procedure – step by step / how ethical issues were addressed / research design and method / sample, sampling method, target population.

Answer 125

Descriptive stats / inferential stats / graphs / level of sig / thematic analysis.

Answer 126

Qualitative summary of results, reflection of results against background research, strengths, limitations, implications, future research.

Answer 127

Details of source materials used

Answer 128

Researchers’ surnames, Initial.. (Year). Name of article. Journal name, Chapter, Page range.

Answer 129

Researchers’ surnames., (Year, Month). Magazine article title. Magazine, Page range.

Answer 130

Author’s name. (Year of publication). Title of book. Location, Publisher.

Answer 131

Data is represented in the form of categories. Nominal data represents the frequency of people or items in each category. Nominal data is discrete in that one item (or participant value) can only be in one of the categories.

Answer 132

Data is ordered in some way. Ordinal data does not have equal intervals between each unit. Psychology questionnaires collect data that is “treated as ordinal” data. The participant provides a self-report of their thoughts and feelings (which is subjective) and often they use a rating scale. The gap between 0-1 might not be equal to the gap between 1-2.

Answer 133

Interval data is based on numerical scales that include units of equal, precisely defined size. These units have equal intervals between them. These are universal scales of measurement that produce data based on accepted units of measurement e.g. time (minutes / seconds), temperature (degrees Celsius) and weight (kilograms). Interval data is the most precise and sophisticated form of data in psychology and is necessary for using parametric tests (e.g. unrelated t-test, related t-test and Pearson’s r)

Answer 134

carrots should come mashed with suede under roast potatoes

Answer 135

directional hypothesis

Answer 136

non-directional hypothesis

Answer 137

the calculated value of () is greater than/less than the critical value () when N=(), for a one/two-tailed test where p<(). Therefore the difference found is/is not significant. This means we accept/reject the null hypothesis

Answer 138

a false positive. incorrectly rejects a null hypothesis. likely to happen if the significance level selected is too lenient e.g. p< 0.01

Answer 139

a false negative. incorrectly accepts the null hypothesis. likely to happen if the significance level selected is too stringent e.g. p<0.001

Answer 140

convert data to nominal. write + or - in last column.

Answer 141

count up all the pluses and all the minuses. ignore any that are the same in condition A and B. lowest value is calculated S value

Answer 142

find he critical value using the critical values table

Answer 143

Consent Deception Confidentiality Debrief Withdraw (Right to) Protection from harm

Answer 144

making participants aware of the aims, procedures, their rights (right to withdraw,) what data will be used for

Answer 145

deliberately misleading or withholding information from participants

Answer 146

should not be placed at any more risk than they would be in their daily lives

Answer 147

set of ethical guidelines that researchers have a professional duty to observe these guidelines

Answer 148

participants should be issued with a consent letter or form detailing all relevant information that might affect decision to participate

Answer 149

- a method of quantifying qualitative content via categories - a form of observation that examines forms of media people produce

Answer 150

categories, tally

Answer 151

qualitative to quantitative

Answer 152

form of content analysis but outcome is qualitative. involves identifying implicit or explicit recurrent, ideas within the data. themes will emerge once data has been coded.

Answer 153

coding, themes

Answer 154

remains qualitative

Answer 155

- no ethical issues as information is widely available - high external validity - flexible in producing quantitative and qualitative data

Answer 156

- lack of objectivity - researcher bias

Answer 157

objectivity, empirical method, replicability, falsifiability, theory construction, hypothesis testing, paradigm, paradigm shift

Answer 158

all sources of personal bias are minimised so as to not distort or influence the research process

Answer 159

scientific approaches that are based on the gathering of evidence through direct observation and experience

Answer 160

the extent to which scientific procedures and findings can be repeated by other researchers

Answer 161

the principle that a theory cannot be considered scientific unless it admits the possibility of being proved untrue

Answer 162

the process of developing an explanation for the cause of behaviour by systematically gathering evidence and then organising this into a coherent account

Answer 163

a key feature of the theory is that it should produce statements which can then be tested

Answer 164

a set of shared assumptions and agreed methods within a scientific discipline

Answer 165

the result of a scientific revolution when there is a significant change in the dominant unifying theory within a scientific discipline