rs statistics Flashcards
(87 cards)
1
Q
What’s quantitative data
A
*data that is expressed numerically
*data can be gained from individual scores in experiments
2
Q
Strengths of using quantitative data
A
*simpler to analyse
*allows comparisons to be drawn between groups of data
*patterns and trends can be established
3
Q
Weakness of using quantitative data
A
*lacks depth and meaning to beahviour
*prevents ppts from being able to develop their thoughts/feelings and opinions
*lack vital data and reduces the internal validity
4
Q
Strengths of qualitative data
A
*provide rich detail and depth
*allows ppts to develop their thoughts and feelings on given subject
*greater understanding of behaviour being studied
4
Q
Whats qualitative data
A
*descriptive data
*may take the form a written description of thoughts, feelings and opinions usch as notes from interview or open questions in questionnare.
5
Q
Weaknesses of qualitative data
A
*harder to analyse as difficult to summarise statistically to establish patterns and trends
*opens data to potential researcher bias as analysis in based upon their subjective interpretations of data
6
Q
What are the diffrent types of quantitative data
A
*Nominal
*Ordinal
*Interval
7
Q
Describe nominal data
A
*data in the form of categories
8
Q
Describe ordinal data
A
*ordered/ranked
*doesnt have fixed intervals
*data based on subjective opinions
9
Q
Whats a weakness of ordinal data
A
*seen as unsafe as it lacks precision and not used part of statistical testing
10
Q
Describe interval data
A
*data is standardised/universal measurement
*based on objectiv emeausres like time in seconds
*based on numerical scales that include units of equal size
11
Q
Whats primary data
A
*data gathered directly from ppts themselves
*specific to the aim of the study
12
Q
How can primary data be gathered?
A
*experiments
*questionnares
*interview
*observations
13
Q
Strengths of primary data
A
*collected first hand from ppts spcifically for thr aim
*allows researchers to specifically target the info that they need
*organise experiment in a way that suits them and their aim
14
Q
Weaknesses of primary data
A
*conducted by the researcher
*involves time and effort to obtrain the data
15
Q
Whats secondary data
A
*data previously collected by a third partuy
*not specifically for aim of the study, used by researcher
16
Q
Purpose of using bar charts and how do u structure it
A
*used to display discrete/categorical data
*used to compare conditions
Struc:
*categories on the x axis
*frequency or amount is on y axis
17
Q
Purpose of histograms and how do u structure it
A
*display continuous data or data where each ppts has individual score
Structure;
*scores on x axis
*the frequency (number of) will be represented on y axis
*bars always touching
18
Q
Purpose of scattergraphs and how do u structure it
A
*used to display a relationship between 2 co-variables
*represent correlations
19
Q
What’s the purpose of statistical tests
A
*determine likelihood that results/difference/correlation are due to chance or the IV/ co-variables
*indicates what hypothesis- null or alt will be more likely to be true- most probable
20
Q
Whats the process of the stats test
A
*carry out stats test and get a calculated value
*compared calc to critical value- decides if calc value is significant.
*if calc value is sig, can accept alt hypothesis
*if not sig- accept null and reject alt hypothesis
21
Q
If the experimental design is repeated/matched and data is nominal what stats test do u use
A
*sign test
22
Q
repeated/matched and data is ordinal what stats test do u use
A
*wilcoxon
23
Q
repeated/matched and data is interval what stats test do u use
A
*related t test
24
independent and nominal- what stats test
*chi- squared
25
independent and ordinal- what stats test
mann whitney u
26
independent and interval- what stats test
unrelated t test
27
correlation and nominal- what stats test
chi squared x2
28
correlation and ordinal- what stats test
spearmans rho
29
correlation and interval- what stats test
pearsons r
30
Explain why a repeated measures design was more appropriate than an independent groups design in this study.
[2 marks]
* a repeated measures design would control for individual differences
* so, the psychologist can be more certain that any difference in DV between the two conditions is IV rather than due to other participant variables.
31
Explain one strength and one limitation of collecting quantitative data in this study.
[4 marks]
* quantitative data is straightforward to analyse so the IV can be easily assessed
* quantitative data is too restrictive to assess the IV as the feeling/type/duration/context of IV
is not given, decreasing the validity of the data
32
Explain one limitation of using the range to represent the spread of stress ratings.
[4 marks]
* these range values only take the two extreme stress ratings into account
* these range values are not calculated using each individual stress rating
* these range values can be distorted by outliers, e.g. it could be the case that most students in condition A had a stress ratings of 4 but one individual hated running and so reported a stress rating of
10 in this condition, this could result in the range of 6 as shown but would not represent the spread of the data collected.
33
Explain what is meant by a type II error in the context of this study.
[2 marks]
whe null hypothesis is wrongly accepted
(when the difference is actually due to the experimental manipulation) – when difference between DV across the conditions is rejected as being not significant when there is a real difference between stress levels after running 2 km as opposed to not running
* a false negative – concluded that there was no significant difference
34
Suggest one way the psychologist could reduce the chance of a type II error.
[1 mark]
* set a less stringent/less demanding significance level
* use the 5% level
* increase the sample size
35
Describe how to use stratified sampling
* identify the strata that make up population. Work out the proportions needed for the sample to be representative. Use random sampling to select the participants that
make up each stratum
36
Briefly explain one reason why research should be peer reviewed
* to reduce chances of flawed/unscientific research being published and therefore treated as ‘fact’ and misinforming the public
* to improve quality of published reports by suggesting amendments or further control/extension work
*dtermines whether rs should receive funding
* would ensure that the research is appropriate for publication, eg suitable terminology and language
used for the target audience and changed if required, check work is original (not copied/plagerised)
* would ensure that the research is credible and thus maintaining the reputation of the institution and
securing future funding for research.
37
Identify two ethical issues in the design and conduct of psychological research.
In each case, explain how the issue you have identified could be dealt with.
[6 marks]
* informed consent – participants should be given a consent form including all the details of the study so
they can make an informed decision as to whether they wish to participate; for children under 16 years
their parent/guardian needs to sign on their behalf; presumptive consent/retrospective consent
* deception – fully debrief participants at the end of the study and make aware of the true aims of the
study; participants should be given the right to withdraw their data from the study
* protection from harm – reassurance provided about their behaviour/performance in the study (debrief);
counselling provided if necessary; fully debrief participants at the end of the study and make aware of
the true aims of the study; participants should be given the right to withdraw their data from the study
* privacy & confidentiality – protect any personal details collected; maintain anonymity by using
pseudonyms/initials/numbers when referring to participants.
38
Explain why the researcher used a controlled observation and a standardised script in this study.
[4 marks]
* a standardised script would ensure the same greeting was given to the children and their parents when they arrived, providing a high level of control
* without using a standardised script, the researcher’s welcome may vary causing unintentional bias, eg may use more child-friendly vocabulary which may make the child less anxious and thus more
sociable, etc
* using a controlled observation helps to reduce extraneous variables which may have an impact on the social behaviour of the children, affecting the validity of the study.
39
Explain why your chosen categories are
appropriate.
* behavioural categories are appropriate because they focus on social behaviours/are objective/visible/mutually exclusive/not ambiguous/not overlapping.
40
Describe how the observers could use time sampling to record the social behaviour of each child during the five-minute period
* record the child’s behaviour at set time intervals during the five-minute period, eg every 20 seconds
* at the specified time intervals, tick/mark one or more categories from the behavioural checklist according to the behaviour exhibited by the observed child, eg if the child is smiling and chatting to
another child the observer should tick ‘smiling at another child’ and ‘talking to another child’ from the list of behaviours
* behaviours observed in between the time sampling frames should be ignored and should not be recorded, eg even if a child is sharing their toy after time sampling snap shot, this behaviour should
not be recorded.
41
Explain one strength and one limitation of using time sampling for this observation.
Possible strengths:
* time sampling may be easier to manage as there are likely to be many social behaviours occurring throughout the set observation time
* time sampling is most appropriate for recording ongoing behaviours that are likely to occur in a social setting, eg conversations or playing together.
Possible limitation:
* time sampling may not be representative of social behaviours as some social behaviours that 3-year-old children may display may be missed,
42
Explain how the reliability of the controlled observation could be assessed through
inter-observer reliability.
[4 marks
* two observers would use same behavioural categories
* two observers would make independent observations/tallies (of the same child at the same time/the 5-minute sessions are filmed and each observer watches and records the data for each film)
* the two observers’ tally charts would be compared to calculate the correlation
between the recordings of the two observers to determine the level of inter-observer reliability
* researchers generally accept +0.8 correlation as a reasonable degree of reliabilit
43
Describe features of the abstract section in a scientific report.
[3 marks
* first section of a report
* brief/summary/100-300 words (accept alternative wording)
* contains (a summary of) the aims, methods, results, and conclusions
44
Explain one strength of collecting quantitative data
* would enable the researcher to easily analyse the data
* would enable the researcher to perform a statistical test to enable conclusions about
45
Explain one reason why the researcher decided to include these additional questions on the questionnaire.
[2 marks]
* additional questions distract the students from realising that the focus of the study was on IV
* students would be less likely to alter their ratings about their sleep quality, reducing demand characteristics (improving validity).
46
Outline one reason why it was appropriate to conduct a correlation rather than an
experiment in this case.
[2 marks]
* it would be difficult to dictate the number of hours the students should sleep
* it may be more ethical to conduct a correlation rather than restrict how many hours each student slept
47
Explain why the researcher decided to use the 5% level of significance rather than the
1% level in this study.
* the 5% level of significance is the conventional level of probability employed by psychologists/balances the risks of making a Type I and Type II error
48
In which section(s) of a scientific report would you expect to find reference to the
results/findings of the investigation?
[1 mark]
The abstract, the discussion and the results sections only
49
What is meant by the term meta-analysis?
* process where researchers collect and collate a wide range of previously conducted research on a specific area
* collated research is reviewed together
* combined data/effect size is often statistically tested to provide an overall conclusion
50
What are investigator effects? Suggest one way in which they could have been minimised
*Any (unintentional) influence of the researcher’s behaviour/characteristics on participants/data/outcome
*Possible ways:
provide a standardised script for the interviewers to use so that they all asked the same questions in the same way to avoid any bias in the students’ responses
the interviewers could have been trained to greet the students in the same way and ask questions with a neutral tone
ensure all interviewers were female or all interviewers were male
ensure that the students were interviewed by someone of the same gender as themselves.
51
Content analysis
identify/devise important categories/themes (from the interviews)
example of possible category/theme (eg references to aggressive interactions)
work through transcribed conversations / repeatedly listen to segments of the recording
count/tally the number of occurrences of each of the categories/themes.
52
Test re test
content analysis repeated on a second occasion using the same x data
compare the results of the two separate analysis (number of occurrences of each)
researchers could calculate the correlation between the two ratings
researchers generally accept 0.8 correlation (accept 0.7-0.9) between the test and the re-test
53
Inter rater
use a second person to work with the original researcher
they could read the interviews (separately) and devise a set of categories (and agree operational definitions)
they could tally the occurrences of each of the categories of the interviews (separately)
they could compare their tally charts looking for agreement
researchers could calculate the correlation between the two ratings
researchers generally accept 0.8 correlation (accept 0.7-0.9) between the test and the re-test
54
Identify one variable on which participants should be matched in this matched-pairs
design. Explain how the researcher could assign matched participants to either Group A or Group B.
1 mark for identifying an appropriate variable that participants could be matched for, eg personality type,
happiness, type of sport they usually participate in.
Plus
1 mark for each of the following:
* explanation of how the participant variable will be measured, eg participants could be paired together
for happiness using the initial happiness questionnaire OR participants could be paired together for
the type/frequency of sport they usually participate in by asking them to complete a sports survey
* participants with the two highest scores are paired, then the next two highest score are paired and so
on until all participants are paired (10 pairs)
* one participant from each pair would be randomly allocated to a different condition of the experiment
(so that there were 10 participants in each group).
55
Explain what it means for a test to have high concurrent validity
* high concurrent validity is where there is close agreement between the data produced by the new test
compared to the established test. Close agreement is indicated if the correlation between the two sets of data produced by the two tests exceeds +0.8
56
Validity was still a concern because the researcher knew which participants were in
each experimental group.
Explain how this could have affected the validity of the study
* knowing which experimental group each participant was in could lead to investigator
effects/researcher bias
* the researcher may have unconsciously/consciously influenced responses given by the participants
eg by smiling more/less or using a more/less pleasant tone/etc. to the participants assigned to one of
the groups whilst they were completing their happiness questionnaire
* this would have made them more/less likely to rate themselves as happier
* the increased happiness in the team sports group may therefore be due to investigator
effects/researcher bias as opposed to the change in the IV invalidating the results of the study
57
suggest one way in which the researcher could modify the study to improve the internal validity of the study. Justify your answer
Possible content:
* they could ensure neither the participants nor the investigators know which condition the participants
are in (double blind procedure). If the researcher did not know what condition the participants were in
they would not be able to consciously/unconsciously give clues
* the participants could have completed the questionnaires without the presence of the researcher by
sending them the questionnaire by post or electronically with instructions provided including how to
return the completed questionnaires. If the researcher was not present whilst the participants carried
out the questionnaire then their presence would not affect the results obtained
* the researcher could have asked an assistant to conduct the experiment who was unaware (blind) to
the conditions the participants were assigned to. If the assistant did not know the conditions to which
the participants were assigned then their presence would not affect the results obtained.
58
What does a negatively skewed distrubition show
more scores are higher end eg showing more ppl did well with outliers at lower end
59
Explain how mode, median and mean are placed in a distribution
mode is the highest point- as most frequent score
mean- shifted towards outlier scores in tail
median- in between mean and mode
60
What side is the tail on a negative skew
left with the distrbutions being saturated on the right
61
What side is the tail on a positive skew
right with distrbutions saturated at left
62
Whats the abstract and what does it include
the start of a scientifc report
*acts a summary- has aim, ,method, sample
*allows researcher to see if article is relevant and if they should read on
63
Whats the purpose of the intro and what does it contain
*says WHY research is being conducted
*explains aim and previous rs in area
*gives rs ability to replicate
64
Whats the purpose of the method
includes sample- procedure- equip- ethics, design
65
What does results show in scientific report
*overall summary of data
*includes descriptive and inferential stats
66
Purpose of discussion
*discusses findings and suggests possible uses and future areas of rs
*limitations and modifications
*compare w other results
67
How do u reference an article
Name- date- tile of article- journal- volume- issue number- page number
68
How do u reference a book
Author's last name- followed by date, title of book. Place of publication: publisher
69
How do u write a consent form and what tense
FUTURE tense
*thank ppts for taking part
*aim and outline of task and how much time will take
*explain ethical issues
*remind that they can ask any questions
70
How do u write standardised instructions and tense
FUTURE
*formal but polite
*you will be required to...
*how long itll take to complete task
*what have to do when finished
*any questions
71
How to write a debrief and what tense
PAST
*thank ppts
*include aim of rs
*if deceived tell them why u had to do it
*if independent groups- tell them what condition they didnt take part in
*explain ethical issues and how been accounted for
*reassure that behaviour is normal and offer care if needed
*ask if any questions
72
Give one disadvantage of using median values as seen in Figure 1 to represent the anger scores of participants.
Possible answers
* Any outlier values/extreme values (such as the scores of 50) would be ignored/would not form
part of the average measurement.
* Less sensitive than the mean.
* It does not represent all the findings
73
Explain how demand characteristics might have occurred in this study using questionnare
* Participants might ‘respond’ with an increase/decrease in anger in order to ‘ruin’ results.
* When the researcher gives out the questionnaire for the second time, this sets up a demand
characteristic that the second answers should be different to the first answers.
74
The researcher used a questionnaire to assess the anger scores of the offenders. Briefly discuss one strength of using questionnaires in research.
the ability to collect large amounts of data relatively quickly and conveniently; discussion of how they
might be easy to score/collate – when items are closed; reference to standardisation/replication of
method.
75
Explain why the median would be more suitable than the mode to represent the average scores for satisfaction for the students.
The median is more representative than the mode
*mode is at the top end of the set of scores,
*whereas more of the scores are numerically closer to the median.
76
Briefly explain one problem the researcher might have in analysing the qualitative data
collected from the questionnaires
analysis of qualitative data relies on interpretation by the researcher which could be inaccurate/biased/subjective
qualitative data may not be easy to categorise/collate into a sensible number of answer types
the researcher could be left with lots of individual responses that cannot be summarised
77
Explain why it was important to match the students on their high level of interest in business studies.
level of interest could affect their satisfaction scores so needs to be controlled; matching students’
levels of interest in business studies keeps this participant variable constant.
78
Briefly discuss one way in which a covert observation of children might be more
beneficial than an overt observation. 3m
* compared to overt in covert children are completely unaware that they are being monitored and this increases the likelihood of the behaviour that is produced being
spontaneous and natural
the researcher can also be more certain that covertly collected data are more likely to be valid.
*measurements of behaviour and/or that these behaviours might therefore indicate possible cause and
effect relationships.
79
What is a matched pairs design?
* participants are matched/paired according to key variables
* each person from a pair goes into a different condition.
80
How is using a matched pairs design an improvement on an independent groups
design?
* this will reduce possible effects of participant variables
* this increases the ability of the researcher to conclude that differences in results are due to different conditions (not different people)
81
Explain why collecting qualitative data in this study might have been a more useful way of assessing the effectiveness of the two therapies.
* qualitative data allows the clients to express fully their reasons for their therapy ratings so the researchers can be more sure that their conclusions about the therapies are both valid and reliable
* the single number ratings about their feelings before and after therapy do not explain in any detail how
the therapy affected the clients
82
Explain the benefit of using randomisation to produce the word lists for this study
* eliminates investigator effect/bias - one word list could not be made to be easier or harder than the other when the lists are constructed
* this is because the investigator has no control/choice over which words appear on which lis
83
The 50 words used to make Word list A and Word list B were not standardised in any
way.
Explain how the words that were used to make the two lists should be matched in order to improve the study
* all the (50) words should be chosen/matched as having equal/same level of difficulty as each other
(accept examples such as same number of syllables/letters/type of word)
84
Given the information above, explain why it was more appropriate to use the standard
deviation rather than the range.
* SD is not easily distorted by extreme scores (such as the anomalous score in Condition 1)
* SD takes into account the distance of each score from the mean, not just take the distance between the highest and lowest scores
* SD uses every piece of data/value collected
* the comparison of the variations in performance would be more accurate
85
Explain one way in which using open questions, in addition to closed questions, might improve the questionnaire.
[2 marks]
* open questions might give the researcher detailed insight into reasons for behaviour
* open questions could lead to ideas for further investigation
* respondents find open questions less frustrating than forced choice.
86
Explain what is meant by operationalisation
clear identification/definition of the observable actions/behaviours to be recorded
* this enables the behaviour under review to be measured objectively.
