Research Methods Flashcards

Question

Outline stratified sampling

Answer 1

Subgroups within a population are identified, for example gender or age, and participants are obtained from each of the strata in proportion to their occurrence in the population. Selection from the strata is done randomly. + Representative - Time consuming

Answer 2

Using a predetermined system to select participants, such as selecting every nth person from a phone book. + Unbiased - Time consuming

Answer 3

+ May be more representative and less biased - Participants more likely to be highly motivated to be helpful or want the reward - Volunteer bias

Answer 4

Informed consent Deception Confidentiality and privacy Debrief Right to withdraw Protection from (psychological) harm

Answer 5

A cost-benefit analysis -> the costs of doing the research is judged against the benefits. The costs and benefits can be judged from the point of view of participants, society as a whole, or in terms of the group to which an individual belongs Ethics committees -> these committees must approve any study beforehand, and the committee looks at all possible ethical issues raised in any research proposal and how the researcher suggests those issues will be dealt with Punishment -> if a psychologist behaves in an unethical manner, the BPS reviews the research and may decide to bar the person from practicing as a psychologist.

Answer 6

In a naturalistic observation, behaviour is studied in a natural situation where everything has been left as it is normally. + Gives a realistic picture of spontaneous behaviour - Little control over variables

Answer 7

In a controlled observation, some variables in the environment are regulated by the researcher, reducing the 'naturalness' of the environment, and the 'naturalness' of the behaviour being studied. + An observer can focus on particular aspects of behaviour - Control of variables means environment and behaviour is less natural

Answer 8

People are observed without their knowledge. + Less demand characteristics - increased validity + Natural behaviour - Ethical issues

Answer 9

Participants are aware their behaviour is being studied. - Demand characteristics - Unnatural behaviour

Answer 10

Observations are made by someone who is also participating in the activity being observed. + Provides special insights into behaviour that might not otherwise be gained - Likely to be overt and thus have issues of participant awareness

Answer 11

The observer is separate from the people being observed. + Observers are likely to be more objective - Likely to be covert, and thus ethical issues may arise

Answer 12

All relevant behaviour is recorded with no system. An issue is there may be too much too record. Furthermore, the behaviours recorded will often be those which are the most visible or eye-catching to the observer, but these may not be the most important or relevant behaviours. Researchers may use this approach where research has not been conducted before, as a kind of pilot study to see what behaviours might be recorded.

Answer 13

The researcher has a system used to record behaviour. Behavioural categories: Behaviour being studied is broken up into a set of components. Behavioural categories should: - Be objective - Cover all possible component behaviours and avoid a 'waste basket' category - Be mutually exclusive Sampling procedures: Event sampling - counting the number of times a certain behaviour occurs Time sampling - recording behaviours in a given tome frame, for example, noting what the participant is doing every 30 seconds

Answer 14

Turning the DV into a measurable format

Answer 15

+ Allows access to what people think and feel, and to their experiences and attitudes - People may not supply truthful answers due to social desirability bias - If people do not know the answer to a question, they often make it up, meaning answers may lack validity - The sample of people used in any study using self-report may lack representativeness and thus the data may lack generalisability

Answer 16

Questionnaires are a self report technique. They allow researchers to discover what people think and feel, in contrast to observations, where researchers have to 'guess' what people think and feel on the basis of how they behave. + Once designed and tested, questionnaires can be distributed to large numbers of people relatively cheaply and quickly, enabling a researcher to collect data from a large sample of people + Respondents may be more willing to give personal information in a questionnaire than an interview, where they may feel self-conscious and more cautious - Questionnaires are only filled in by people who can read and write and have the time to fill them in, meaning the sample is biased

Answer 17

A structured interview has pre-determined questions, so is essentially a questionnaire that is delivered face to face. It is a self-report technique. + The main strength of a structured interview is that it can be easily repeated because the questions are standardised, meaning answers from different people can be compared + This also means that they are easier to analyse there an unstructured interview because answers are more predictable - Comparability may be a problem in a structured interview, if the same interviewer behaves differently on different occasions or different interviewers behave differently - A limitation of both structured and unstructured interviews is that the interviewers expectations may influence the answers the interviewee gives - interviewer bias

Answer 18

In an unstructured interview new questions are developed during the course of the interview. It is a self-report technique. + More detailed information can generally be obtained from each respondent than in a structured interview because the interviewer tailors further questions to the specific response - Unstructured interviews require interviewers with more skill than a structured interview because the interviewer has to develop new questions on the spot - Such in-depth questions may be more likely to lack objectivity than predetermined ones, because of their instantaneous nature, with no time for the interviewer to reflect on what to say

Answer 19

There are three guiding principles when writing questions: 1. Clarity -> there should be no ambiguity in questions 2. Bias -> any bias in a question may lead the respondent to be more likely to give a particular answer, with the biggest issue being social desirability bias 3. Analysis -> questions need to be written so the answers are easy to analyse, so closed questions would be better to analyse

Answer 20

A research method that investigates the relationship between 2 co-variables.

Answer 21

Things examined together for a relationship. The co-variables are both measured.

Answer 22

- A large-scale review study where a researcher takes the data of lots of individual studies and then analyses it all together to get an overall picture - It’s an ‘analysis of analyses’

Answer 23

Mean: + The most sensitive measure of central tendency because it takes account of the exact distance between all the values of all the data - Can be easily distorted by an anomaly so may be misrepresentative of the data as a whole - Cannot be used with nominal data Median: + Not affected by extreme scores + Easier to calculate than the mean - Not as 'sensitive' as the mean because the exact values are not reflected in the final calculation Mode: + Unaffected by extreme values + Only method that can be used on nominal data - Not a useful way of describing data when there are several modes - Tells us nothing about the other values in a distribution

Answer 24

Range: + Easy to calculate - Affected by extreme values - Fails to take into account the distribution of the numbers Standard deviation: + A precise measure of dispersion because it takes all the exact values into account - May hide some of the characteristics of the data set, such as extreme values

Answer 25

-3 = 0.13% -2 = 2.15% -1 = 13.59% 0 = 68.26% +1= 13.59%. +2 = 2.15% +3 =0.13%

Answer 26

Positively skewed distribution - right sided skew, goes mode median mean - high to low Negatively skewed distribution - left sided skew, goes mean median mode - low to high

Answer 27

Numerical data + easy to analyse, test with statistics, and compare for patterns and trends + objective, scientific, less open to investigator bias - not detailed, doesn't explain the reasons behind behaviour

Answer 28

Non-numerical data + detailed and can provide real insight into why behaviour is the way it is, behaviour isn't limited by numbers - anything can be recorded - more subjective and open to investigator bias - nor easy to analyse and compare for patterns and trends

Answer 29

Information observed or collected directly from firsthand experience + a strength of generating primary data is the researcher has full control over the data. The data collection can be designed to fit the aims and hypothesis of the study. - it is a very lengthy and therefore expensive process. Simply designing a study takes a lot of time and then time is spent recruiting participants, conducting the study, and analysing the data.

Answer 30

Information that was collected for a purpose other than the current one + it is simpler and cheaper to just access someone else’s data because significantly less time and equipment is needed + such data may have been subjected to statistical testing and thus it is known whether it is significant - the data may not exactly fit the needs of the study

Answer 31

Nominal Ordinal Interval

Answer 32

- This type of data is used to classify or categorise things like yes/no male/female answers when collating data in a study - This system is generally accepted as a means of identification - A set of data is said to be nominal if the values/observations belonging to it can be assigned a code in the form of a number where the numbers are simply labels - You can count but not order or measure nominal data

Answer 33

- A set of data is said to be ordinal if the values/observations belonging to it can be ranked or have a rating scale attached - You can count and order, but not measure ordinal data

Answer 34

- Data is measured in using units of equal intervals - Scales are excepted precise ones, such as distance, time, and temperature - With this scale, arithmetic operations are possible - this is because the numbers are ordered and the interval between each number along the scale is of equal size

Answer 35

[Test of difference] [Correlation] [Independent] [Repeated and matched] [Correlation] Nom: [Chi Squared] [Sign Test] [Chi Squared] Ord: [Mann-Whitney] [Wilcoxon] [Spearman's Rho] Int: [Unrelated T] [Related T] [Pearson's R] - If the name of the test has an R in it, the observed value needs to be moRe than the critical value - If the name of the test doesn't have an R in it, the observed value needs to be less than the critical value - Statistical tests for nominal and ordinal data are non-parametric Statistical tests for interval data are parametric

Answer 36

- A false positive - Where the null hypothesis is rejected, and the alternative hypothesis is accepted - Where the researchers claim to have found a significant difference when there is not one

Answer 37

- A false negative - The null hypothesis is accepted, and the alternative hypothesis is rejected - The claim of no significant difference between variables, when there actually is one

Answer 38

The null hypothesis is written alongside the experimental one, and states there is 'no difference' between the conditions. The results of the experiment will determine which hypothesis is 'true', and which one is rejected.

Answer 39

Allocation of funding Publication of research in academic journals and books Assessing the research rating of university departments

Answer 40

In the early 1950s, Cyril Burt published results from a twin study that was used to show intelligence is inherited. He started with 21 pairs of twins raised apart and then increased that to 42. The suspicious consistency of the correlation coefficient led to him being accused of inventing data. His two research assistants didn’t exist. However, his research was used to shape social policy. He helped establish the 11+ exams to identify who should go to grammar schools. He argued that since IQ was largely inherited, it was appropriate to segregate children into schools suited to their ability.

Answer 41

There is an extremely large amount of information available on the Internet. Scientific information is available on numerous online blogs, online journals, and Wikipedia. Information is often policed by the ‘wisdom of crowds’ approach. Readers decide whether it is valid or not and post comments. On Philica, papers are ranked on the basis of peer reviews which can be read by anyone.

Answer 42

A limitation is it isn’t always possible to find an appropriate expert to review a research proposal or report. This means that poor research may be passed because the reviewer didn’t really understand it. In addition, reviewers may be bias towards prestigious researchers rather than less well-known names. This emphasises the need to always take a rigourous and questioning approach both to the appointment of reviewers and into their work. A limitation of peer review is publication bias. Journals prefer to publish positive results because editors want research that has important implications to increase the standing of the journal. This results in a bias in published research that leads to misperception of facts. It also appears that journals avoid publishing replications. Ritchie et al submitted a replication of a study on paranormal phenomena and found that it wasn’t even considered for peer review. This suggests that academic journals are as bad as newspapers for seeking eye catching stories. With peer reviews, reviewers have to keep their identity secret. This aims to allow them to be honest and objective. However, it may have the opposite effect; reviewers may use the veil of anonymity to settle old scores or bury rival research. Research is conducted in a social world where people compete for research grants and jobs and make friends and enemies. Social relationships inevitably affect objectivity. To combat these issues, some journals now favour open reviewing, where both author and reviewer know each other’s identity.

Answer 43

Abstract Introduction Method Results Discussion Referencing

Answer 44

- short summary of all sections of the report (150-200 words) - includes aims / hypothesis, method, results, and conclusion - it is useful for psychologists to decide which areas of interest are worthy of future investigation

Answer 45

- a discussion of the general area of interest of your study - includes findings from other major / relevant research in the area - should begin quite broadly, but then become more specific in relation to your area of interest - the final part is where you present your aims and hypothesis

Answer 46

- must include sufficient detail so other researchers can replicate the study if they wish It outlines: - design, both experimental and actual method - sample - apparatus / material - procedure - ethics

Answer 47

- should summarise key findings - descriptive statistics, such as graphs and tables - inferential statistics, such as statistical tests - raw data - if qualitative data has been used then a thematic analysis

Answer 48

- summary of findings in written rather than statistical form - discussed in the context of the introduction - should acknowledge weaknesses and suggests how these may be addressed in a future study - consideration of the wider implications of the research

Answer 49

1. Objectivity and the empirical method 2. Replicability 3. Falsifiability 4. Theory construction and hypothesis testing 5. Paradigms and paradigm shifts

Answer 50

- scientists aim to be objective in their investigations - this means they keep a 'critical distance' in their research - they must not allow their personal opinions or biases to 'discolour' the data they collect or influence the behaviour of the participants they are studying - objective methods in psychology are usually those where there is high control, such as lab experiments

Answer 51

- objectivity is the basis of the empirical method - empirical methods emphasise the importance of data collection based on direct, sensory experience - the experimental and observational methods are good examples of the empirical method in psychology - a theory cannot claim to be scientific unless it has been empirically tested and verified

Answer 52

- replication has an important role in determining the validity of a finding - if a scientific theory is to be 'trusted', the findings from it must be shown to be repeatable across a number of different contexts and circumstances - replicating findings over different contexts and circumstances allows us to see the extent to which findings can be generalised

Answer 53

- in the 1930s, Karl Popper asserted that the key criterion of a scientific theory is falsifiability - genuine scientific theories, Popper suggested, should hold themselves up for hypothesis testing and the possibility of being proved false - he believed that even when a scientific principle had been successfully and repeatedly tested, it was not necessarily true, but had simply not been proven false - Popper drew a clear line between good science, in which theories are constantly challenged, and what he calle 'pseudosciences' which couldn't be falsified - those scientific theories that survive most attempts to falsify them become the strongest, as they have defied attempts to prove them false - this explains why psychologists avoid using phrases such as 'this proves' on favour of 'this supports' or 'this suggests' - it also explains why alternative hypotheses must also have null hypotheses

Answer 54

- a theory is a set of general laws or principles that can explain events or behaviours - theory construction occurs through gathering evidence via direct observation, using the empirical method - it should be possible to make clear and precise predictions on the basis of a theory - an essential component of a theory is that it can be scientifically tested - theories should suggest a number of possible hypotheses - any one of these hypotheses can then be tested using systematic and objective methods to determine whether it will be supported or refuted - in the case of supporting, the theory is strengthened - in the case of refuting, the theory will be revised, revisited, or rejected - the process of deriving ney hypotheses from an existing theory is known as deduction

Answer 55

Paradigms: - a paradigm is a clear, distinct concept accepted by most people in a scientific field, for example, evolution, plate tectonics, and planets orbiting the sun - Thomas Kuhn suggested that what distinguishes scientific disciplines and non-scientific disciplines is a shared set of assumptions and methods, which is a paradigm - Kuhn suggested that psychology lacks a universally accepted program, and has too many internal disagreements and conflicting approaches, so is best seen as a 'pre-science' Paradigm shifts: - according to Kuhn, progress within a particular science occurs when there is a scientific revolution, which is when a group of researchers begin to question the accepted paradigm, as there is too much contradictory evidence to ignore - Kuhn would argue that psychology has not undergone any paradigm shifts

Answer 56

The extent to which the subject matter within the confines of the research is being measured consistently by the method / procedure or tool. This can be influenced by standardised procedures, standardised instructions, operationalisation of variables, and the tool used to measure the dependent variable.

Answer 57

The extent to which the subject matter is measured consistently outside of the original study, by using the same method / procedure or tool, on the same people, and on subsequent replications. This can be influenced by how quickly the materials used in the methods / the tool date, for example changes of fashions, language, technology, and understanding over time, or if the materials used in the method were used in exceptional circumstances.

Answer 58

1. If a procedure does not have reliability then it means that something has not been done consistently in the study. 2. The measuring tool does not consistently measure the same when used on repeat occurrences. 3. In an observation, researchers recording data could be inconsistent with their observations and what they write down.

Answer 59

1. Test-Retest 2. Inter-Observer Reliability

Answer 60

This checks the reliability of a measuring tool. It is done by administering the same test to the same people on different occasions. The timing is important, as it cannot be too close that they remember the test but not too long that their performance could have genuinely changed. If the scores are similar on both occasions then the tool has reliability. This is checked by correlating their scores normally. If the tool comes out as showing poor reliability, the questionnaire might need changing, as the questions may be too subjective or ambiguous.

Answer 61

This checks that observers assessing behaviour are consistent / reliable with one another. It is done by giving the two (or more) observers the same coding sheet but independently record data from the same participant. If the scores of both observers are similar / correlate well then it would mean they are observing reliably. If the test comes out showing poor reliability, the behavioural categories would need checking, as maybe they are too similar etc, and the observers might need more training. When checking the reliability, data is usually correlated to check consistency. If the correlation coefficient is R+0.8 or above then the study is said to have reliability.

Answer 62

- Lab based experiments should have good control over the procedures - Standardisation

Answer 63

- Use controlled and structured observations - Categories should be measurable and clear - More than one observer should be used, and an inter-observer reliability check should take place - Behaviour should be filmed

Answer 64

- Check the measuring tools and co-variables for reliability using test-retest

Answer 65

- Use test-retest - Anything that is unreliable will need adapting, for example removing questions or making them clearer - Using closed questions rather than open

Answer 66

- Having the same interviewer for each participant - Using a structured interview can improve reliability

Answer 67

- An in-depth investigation of one person or a small group - They use a range of sources, such as the individual, family, and friends - They use many techniques, such as psychological tests, observations, interviews, and experiments - Case studies produce rich, meaningful, descriptive detail about what they are studying - They produce qualitative data

Answer 68

- Rich, in-depth, qualitative data can be collected - A complex interaction of factors can be studied, in contrast to experiments where only one is usually studied

Answer 69

- They are difficult to generalise as each person has unique factors - They are subjective, and may be affected by investigator bias - They are difficult to replicate, which affects the reliability of results

Answer 70

Validity: The validity in case studies will often depend upon the methods being used as part of the study. The validity of of each method, such as observations and interviews, will be assessed separately. Reliability: Case studies often collect qualitative data, which can be open to researcher bias. One way to overcome this is to use inter-observer reliability. By showing the results to another researcher and comparing the results, we can assess how reliable the data is.

Answer 71

- A form of observation, but instead of people studied, communication is studied, such as newspapers, tv ads, magazines etc - It is the analysis of something created by humans - Things are studied for patterns and trends within them and then summarised in a systematic way so conclusions can be made - They can be quantitative or qualitative in their design - The aim is to summarise and describe this communication in a systematic way so overall conclusions can be drawn

Answer 72

Thematic analysis: - A form of content analysis but the outcome is qualitative - It involves organising, describing, and interpreting data - Identified themes become the categories for analysis, with thematic analysis performed through the process of coding - A theme is any idea, implicit or explicit, that is recurring in the data

Answer 73

1. Familiarisation with the data -> get to know the content of the data in detail 2. Coding -> identify important features of the data 3. Searching for themes -> identify patterns of meaning 4. Reviewing themes -> ascertaining whether the themes explain the data and answer the research question 5. Defining and naming themes -> a detailed analysis of each theme 6. Writing up

Answer 74

Coding: - It involves the quantification of qualitative material; the numerical analysis of written, verbal, and visual communications - This creates behavioural categories that will be examined in the objects of study Example coding units: Word -> number of slang words used Theme -> amount of violence on TV Character -> the number of female commentators there are in TV sports programmes Time and space -> the amount of time (on TV) and space (in newspapers) dedicated to eating disorders

Answer 75

Matthews (2012): - He analysed 1200 instances of graffiti from toilet walls in US bars - Graffiti was coded according to categories: Sexual Socio-Political (race, politics, religion) Entertainment (music/film) Physical presence (someone's name) Love/romance Matthews found that males composed more physical presence or sexual graffiti, and females were more romantic in their graffiti.

Answer 76

1. Sampling -> decide on what will be sampled 2. Familiarise -> read over the sample 3. Construct categories 4. Collect data 5. Draw conclusions

Answer 77

- No social desirability / demand characteristics as using something that is already printed, so improves the internal validity. - Content analysis is based on real-life sources of information rather than the artificial situations that are found in experiments. This gives content analysis high mundane realism which also leads to high levels of ecological validity. - Content analyses can be easily replicated, as other researchers can use the same sources and coding systems and find the same results, making it easy to check the reliability of the data.

Answer 78

- It is time consuming to carry out and difficult to do, which can result in issues with incomplete analysis, which affects the internal validity. - There are issues with observer bias, which affects internal validity. - Cause and effect relationships cannot be demonstrated using content analysis, as patterns and trends can be identified, but they could be either reflecting the social world, or causing differences in the social world. - Content analysis may be subject to researcher bias, which is where researchers find what they are looking for because of their preconceived and subjective ideas and beliefs, and this lowers the objectivity of the method.

Answer 79

Validity -> use a double blind Reliability -> have more than one observer analyse the sample, use the same coding sheet / behavioural categories

Answer 80

Face validity: Whether a test, scale, or measure appears 'on the face of it' to measure what it is supposed to measure. This can be determined by simply 'eyeballing' the measuring instrument or by passing it to an expert to check. Concurrent validity: The concurrent validity of a particular test or scale is demonstrated when the results obtained are very close too, or match , those obtained on another recognised and well-established test. Close agreement between the two sets of data indicates that the new test has high concurrent validity, and close agreement is indicated if the correlation between the two sets of scores exceeds +0.8

Answer 81

- Control group - Standardised procedures - Single and double blind procedures

Answer 82

- Lie scale - Anonymous data

Answer 83

- Covert observations - Clear behavioural categories

Answer 84

- Coherence of the researcher's narrative and the inclusion of direct quotes from participants within the report - Triangulation, which is the use of a number of different sources as evidence, for example, data compiled through interviews with friends and family, personal diaries, observations etc

Research Methods Flashcards

(109 cards)