research methods Flashcards

Question

weakness of field experiments

Answer 1

- Loss of control of CV's and EV's - cause and effect between IV and DV is more difficult to establish and precise replication is often not possible - Ethical issues - if participants aren't aware they're being studied they can't consent to being studied therfore research may be considered a breach of privacy

Answer 2

- Provide opportunities for research that may not otherwise be undertaken for practical or ethical reasons - Often have a higher external validity because they involve the study of real world issues and problems as they happen eg. Effects of a natural disaster on stress levels

Answer 3

- Naturally occuring event may only happen very rarely, reducing opportunities for research - may limit the scope for generalising findings to other similar situation - Participants may not be randomly allocated to experimental conditions, means the researcher might be less sure whether the IV affected the DV - Research may be conducted in a lab therefore lacking realism and demand characteristics may be an issue

Answer 4

- Often carried out under controlled conditions therfore share strengths with lab experiments like replication and controlling CVs and EVs Can study naturally occurring differences between participants

Answer 5

- Cannot randomly allocate participants to conditions and therfore there may be confounding variables - IV is not deliberately changed by the researcher and therefore we cannot claim that IV has caused any observed change

Answer 6

Definition of experimental method is the manipulation of the IV to measure the effect it has on the DV however in quasi-experiments and natural experiment the researcher doesn't deliberately change in the IV and has low control(?) over it so it is hard to see if the change in the IV causes a change in results of the DV.

Answer 7

- when we can see cause and effect, no extraneous/confounding variables - whether the effects observed in an experiment are due to the manipulation of the independent variable and not another factor

Answer 8

It refers to how realistic the task is at representing a real experience of the dependent variable.

Answer 9

the extent to which a study measures what it intends to measure

Answer 10

efers to how well you can generalise from research participants (apply the findings of a study) to people, places and times outside of the study

Answer 11

the extent to which behaviours observed and recorded in a study reflect the behaviours that actually occur in the real world.

Answer 12

the extent to which the results of the research can be generalised to people outside of the study

Answer 13

This is the extent to which the results of the research that took place at a certain point in time accurately reflect the way that behaviour would occur at a different point in time.

Answer 14

- mudane realism of task - mudane realism of environment - participants awarness of being studied - generla tule of thumb 2 out of the 3 ave to be covered fo rit to be ecologicaly valid

Answer 15

The different ways in which the participants are assigned to different groups/conditions of the experiment – independent groups, repeated measures or matched pairs

Answer 16

Participants are allocated to different groups where each group represents different experimental conditons, if there are two levels of IV all participants experience one level of the IV only

Answer 17

Pairs of participants are first matched on some variables that may affect the dependent variable. Then one member of the pair is randomly assigned condition A and the other to Condition B - attempt to find the confounding variable of participant variables

Answer 18

All participants take in all conditons of the experiment - two mean scores from both conditions would be compared to see if there was a difference

Answer 19

An attempt to control for participants variables in an independent groups design which ensures that each participant has the same chance of being in one condition as any other

Answer 20

An attempt to control for the effects of order in a repeated measures design: half the participants experience the conditions in one order, and the other half in the opposite order

Answer 21

- the order that a participant takes part in the conditions of a study impacts the results -

Answer 22

- Can use the same materials in both groups so it's standardised - Order effects are unlikely to be an issue because participants only take part in one issue - Demand characteristics are unlikely to be an issue because participants only take part in one condition

Answer 23

- Participant variables are a problem because we have different people in each conditon - More participants needed than for repeated measures so it's less economical (weaker idea) - Random allocation - each participant has an equal chance of ending up in the same condition

Answer 24

- Can use the same materials in both groups so it's standardised - Order effects are unlikely to be an issue because participants only take part in one issue - Demand characteristics are unlikely to be an issue because participants only take part in one condition Some participants variables are controlled through matching

Answer 25

- Participant variables are a problem because we are unlikely to match participants on all possible participant variable - More participants needed than for repeated measures so it's less economical (weaker idea) - Matching is time consuming therefore expensive - Match them on fewer variables using some software after the pre-test (3) Pilot study to work out the most important participant variables and match participants on those

Answer 26

- Participant variables aren't a problem because we have the same people in each conditon - Fewer participants needed than for repeated measures so more economical

Answer 27

- Can't use the same materials in both groups so not standardised - Order effects are likely to be an issue because participants take part in both conditions - Demand characteristics are likely to be an issue because participants take part in both conditions - Counterbalancing - randomly divide the participants into two groups, one group does A then B; other groups do B then A - Time gap between conditions of at least 24 hours - Use different materials but equivalent (equally difficult) materials

Answer 28

The method used to select people from population

Answer 29

A group of people who take part in a research investigation. The sample is drawn from a (target) population and is presumed to be representative of that population i.e stands 'fairly' for the population being studied

Answer 30

A group of poeple who are the focus of the researcher's interest, from which a smaller sample is drawn

Answer 31

Involves participants selecting themselves to be part of a sample (self-selection), researcher may place adverts in newspapers or on noticeboards

Answer 32

Selecting anyone who happens to be willing and available, researcher takes the chance to ask whoever is around at the time of their study eg in the street

Answer 33

Sampling in which al the participants have an equal chance of being selected, the actual sample is selected through the use of some lottery method

Answer 34

Sampling in which the composition of the sample reflects the proportions of people in certain subgroups (strata) within the target population or wider population, researcher

Answer 35

When every nth member of the target population is selected eg every 3rd house on a street, sampling frame produced which is a list of people in a target population then a sampling system is nominated then researcher works through the sampling frame until the sample complete - may begin from a randomly determined start to reduce bias

Answer 36

When certain groups are over or under represented within the sample selected, limits to which extent to whcih generalisation can be made to the target population

Answer 37

Extent to which findings and conclusions from a particular investigation can be broadly applied to the population, possible if the sample of participants is representative of the target population

Answer 38

1. Identify the strata (subgroup) in the target population 2. Calculate the proportions of the target population that each strata represents 3. Randomly select the number of people needed from each strata seperately so the sample represents these proportions

Answer 39

- the lottery method - random number table - random number generator

Answer 40

1. Obtain a list of all the people in the population e.g. the names of all of the people in your school. 2. Put all of the names in a lottery barrel or hat. 3. Select the number of names required.

Answer 41

An alternative random technique is to use a printed table of random numbers. 1. This time every member of the population is given a number. 2. The starting position in the table is determined blindly by placing your finger anywhere. 3. If your population is less than 100, you only need two digit numbers so read the table two digits at a time. If you come to a number that is not in your population (e.g. you have a population of 80 and one number is 93, you ignore the 93 and move on).

Answer 42

Calculators have functions that generate random numbers as do computers and apps on phones. 1. Number every member of the population. 2. Using, for example, Microsoft Excel type =RANDBETWEEN ([lowest number], [highest number]) e.g. if you have a population between 1 and 100, you would type =RANDBETWEEN (1,100)

Answer 43

Any factor that can vary or change within an investigation. They are generally used in experiments to determine if changes in one factor result in changes to another

Answer 44

Some aspect of the experimental situation that is manipulated (changed) by the researcher, or changes naturally, so the effect on the dependent variable can be measured

Answer 45

The variable that is measured by the researcher. Any effect on the dependent variable should be caused by the change in the independent variable

Answer 46

A general statement of what the researcher intends to investigate (the purpose of the study). It is stated at the outset of the study.

Answer 47

The group/condition in the experiment that received the experimental treatment (the independent variable)

Answer 48

The group/condition in the experiment that receives no treatment (they are the baseline)

Answer 49

Clearly specifying/defining observable behaviours that represent the more general construct under investigation/to enable the behaviour under investigation to be measured. In simple terms, this means: clearly defining variables in terms of how they can be measured

Answer 50

There are three steps to writing a hypothesis: 1. Identify both conditions of the independent variable, and the dependent variable. 2. Explain how we could operationalise these variables. 3. Explain how you think the independent variable will impact the dependent variable.

Answer 51

Directional – when there is previous research so that you can predict the direction of the results (you know exactly what to expect). MORE/LESS/SUPERLATIVES Non-directional – when there is no previous research on your topic OR when there is previous research but the studies have found conflicting results so that you cannot predict the direction of the results (you don't know exactly what to expect). For exam questions, you must say what previous research found to justify a directional hypothesis in order to get a mark. DIFFERENCE Null hypotheses This is the opposite of an alternative hypothesis. It is when you predict that there will be no difference or association between the variables that you are studying. In Psychology, research is conducted to try to prove this hypothesis wrong. NO DIFFERENCE

Answer 52

When a conflict or dilemma exists between participants' rights and researchers' needs to gain valuable and meaningful findings, the conflict has implication for the safety and well-being of participants

Answer 53

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

Answer 54

Deliberately misleading or withholding information from participants at any stage of the investigation, occasions where deception can be justified if it doesn't give the participants undue distress

Answer 55

Participants should be protected from physical and psychological harm as they shouldn't be placed in any more risk than their daily lives, this includes reminding participants they can withdraw from the investigation at any point

Answer 56

Our right, enshrined in law under the Data Protection Act, to have any personal data protected

Answer 57

Participants have the right to control information about themselves, right to privacy extends to area where the study took place such that institutions or geographical location are not named

Answer 58

An ethical issue participants should have the right to withdraw from participating in a research study if they are uncomfortable with the study, they should be given contact details so that they can withdraw I.e. to leave at any time and remove their data

Answer 59

A post-research interview designed to inform the participants of the true nature of the study and to restore them to the state they were in at the start of the study

Answer 60

Rather than getting consent fromt the participants themselves a similar group of people are asked if the study is acceptable, if the group agrees then consent of the original participants is presumed

Answer 61

Participants give their permission to take part in a number of different studies including one that will involve deception, by consenting participants are effectively consenting to be deceived

Answer 62

Participants are asked for their consent (during debriefing) having already taken part in the study, they may not have been aware of their participation or they may have been subject to deception

Answer 63

The role of ethics committees to make judgements about the costs and benefits involved in carrying out individual pieces of research

Answer 64

Consent form - tell them the aim, procedure, their rights and what you'll do with their results before the study and get them to sign it if they agree to participate. If they're under 16, you get parental consent Presumptive consent Prior general consent Retrospective consent

Answer 65

Debriefing - tell them the aim, procedure, anything you hid from them (eg. What the other condition was doing), remind them of their rights after the study has occured. Give them the right to withdraw their data

Answer 66

Repeatedly tell them (on the consent form, on the debrief sheet, and tell them repeatedly during the study)

Answer 67

Offer counselling Offer them the right to withdraw Reassure them their behaviour in the study was normal Make sure their is no greater harm than they would experience in everyday life

Answer 68

Assign them a number or use initials or a pseudonym eg. patient X - remind them of this on the consent form and the debrief sheet

Answer 69

A consent form should be appropriate to the target population (e.g. written using age-appropriate language) and should aim to include: 1. The title of the study. 2. An introduction to the purpose of the study with brief details of the background too. 3. Answers to FAQs (which can be written as questions followed by answers, but works better as continuous prose) such as: a. Do I have to take part? b. What will happen to me if I take part? c. How long will it take? d. Where will it take place? e. Are there any risks of taking part? f. Are there any advantages of taking part? g. Who will see the results? Will they be confidential/anonymous? 4. The name and contact details of the researcher (email or phone number). 5. A thank you to the participant for considering taking part in the study. At the end of the consent form, there are a series of statements. The participant confirms their understanding and agreement by ticking a box alongside each statement and signing the form. The statements should include: 1. Confirmation that the participant: a. Has read and understood the information sheet b. Has had the opportunity to ask questions c. Understands that their participation is voluntary and that they are able to withdraw without giving a reason. d. Consents to specific aspects of the study, e.g. being videoed. 2. Their signature and the date.

Answer 70

It should be written so that it can be read out to the participant and include: ◦ True aim of the study ◦ If an independent groups design has been used, there should be an outline of both conditions of the study. Otherwise, an outline of the study should be given. ◦ Ask the participants if they have any questions and if so who they can address these to ◦ Relevant ethical considerations e.g. remind them of their right to withdraw, privacy, confidentiality etc. ◦ Offer someone that they can talk to if their distressed ◦ Thank the participants for taking part

Answer 71

A small scale version of an investigation (involving a handful of participants) that takes place before the real investigation is conducted. The aim is to check that procedures, materials etc work. The aim is to allow the researcher to make changes or modifications if necessary

Answer 72

- To check the procedures, materials etc work and that the investigation runs smoothly - Helpful to find questions in advance and remove or reward those that are ambiguous or confusing - eg. When using self-report methods like questionnaire or interviews - In observational studies provides a way of checking coding systems before the real investigation is undertaken, may be an important part of training observers - Overall, allows the researcher to identify any potential issues and modify the design or procedure saving time and money

Answer 73

A mathematical technique which a researcher investigates a relationship between two variables, called co-variables

Answer 74

As one co-variable increases so does the other eg. number of people in a room and noise tend to be positively corelated

Answer 75

As one co-variable increases the other decreases eg. The number of people in a room and amount of personal we space tend to be negatively correlated

Answer 76

When there is no relationship between the co-variables

Answer 77

The variables investigated within a correlation, for example height and weight, not reffered to as independent and dependent variables because a correlation investigates the association between variables, rather than trying to show a cause-and-effect relationshis

Answer 78

type of relationship between two co-variables where as one variable increases, so does the other variable, but only up to a certain point, after which, as one variable continues to increase, the other decreases.

Answer 79

A statistic to measure the strength of a correlation (the relationship between two or more variables). A correlation coefficient can range between -1.0 (perfect negative) and +1.0 (perfect positive)

Answer 80

In an experiment the researcher manipulates the IV in order to measure the effect in the DV therefore it is possible to infer that deliberate change in the IV caused observable changes in the DV - cause and effect In a correlation there is no manipulation of one variable therefore not possible to establish cause and effect between one co-variable and another, even if there is a strong positive correlation between the two variables you can' assume that one variable caused change in the other

Answer 81

- Useful preliminary tool for research, by assessing strength and direction of a relationship they provide a precise and quantifiable measure of how two variables are related, may suggest ideas for possible future research if variables strongly related or demonstrate interesting pattern - Relatively quick and economical to carry out, no need for a controlled environment and no manipulation of variables is required, data collected by others (secondary data) can be used so correlations are less time-consuming than experiments

Answer 82

- Lack of manipulation and control within as correlation studies can only tell us how variables are related but not why, correlations can't demonstrate cause and effect between variables and therefore we do not know which co-variables is causing the other to change - An untested variable may be causing the relationship between the co-variables we are interested in (intervening variable) - Correlation can occasionally be misused or misinterpreted, relationships between variables are sometimes presented as casual when they aren't especially from the media

Answer 83

1. Type of correlation – positive, negative or zero. 2. Strength of correlation – strong, moderate, weak, zero, moderately strong etc. 3. What the correlation means in plain English – e.g. as height increases, weight increases. 4. Any anomalies or a change in direction.

Answer 84

1. Alternative hypotheses – predict that something will happen. a) Directional (one-tailed) hypotheses – predict one particular outcome e.g. there will be a positive correlation between height and weight. Use when there is previous research suggesting this result. b) Non-directional (two-tailed hypotheses) – predict something will happen but no particular outcome e.g. there will be a correlation between height and weight. Use when there is no previous research or the previous research conflicts. 2. Null hypotheses – predict that nothing will happen e.g. there will be no correlation between height and weight.

Answer 85

results that can be counted, usually given as numbers

Answer 86

results that are expressed in words and non-numerical. They may take the form of a written description of the thoughts, feelings and opinions of participants, or a written account of what the researcher saw in an observation.

Answer 87

data that is based on standardised numerical scales where the units are of equal, precisely defined size (i.e. There are fixed intervals between each unit) e.g. time, temperature, weight.

Answer 88

data that is on a numerical scale where the results can be ordered in some way but the units are not of an equal, precisely defined size (i.e. It is a non-standardised scale where there are not fixed intervals between each unit). This is most commonly results on a questionnaire. For example, if someone gives Psychology 8 out of 10 and another gives it 4 out of 10, it doesn’t make sense to say that the first person likes Psychology twice as much. It is also used when ranking participants e.g. Highest score gets a rank of 1st, next highest gets 2nd, next highest gets 3rd etc. But it may be that 1st won by a very long way but there was only a small difference in score between 2nd and 3rd.

Answer 89

data that is in categories e.g. People choosing which is their favourite colour from red, yellow or blue. You'd have one category of people who choose red, one who choose yellow and one who choose blue

Answer 90

Descriptive statistics are a way of using numbers to describe the data that you have. They don’t tell us whether our results are significant or not (i.e. allow us to know which hypothesis should be accepted or rejected), but they allow us to describe the data in different ways

Answer 91

Measure of central tendency tell us about the central (middle) values for a set of data whereas measures of dispersion tell us how dispersed or spread out the data items are - mean - median - mode

Answer 92

- range - standard deviation

Answer 93

SD is a measure of the average distance between each data item above and below the mean It is more powerful than range because is a more precise way of measuring the spread of data since it shows you the range at each point

Answer 94

Nominal - Mode Ordinal - Median - Range Interval - Mean - SD

Answer 95

This measure of central tendency is not affected by extreme scores so can be useful under such circumstances. This measure of central tendency is appropriate for ordinal data and is easier to calculate than the mean. - This is not as sensitive a measure of central tendency because the exact values are not reflected in the middle value.

Answer 96

This measure of central tendency is the easiest to calculate but is a very crude measure. It can end up being very different from the median and mean and so not really representative of the data as a whole.

Answer 97

-This measure of dispersion is easy to calculate. This measure of dispersion is affected by extreme values because it only takes into account two values. Therefore it may be misrepresentative of the data set as a whole.

Answer 98

- This is a precise measure of dispersion because it takes all of the exact values into account, and so is more likely to be representative of the data as a whole. This measure of dispersion may be distorted by a single extreme value as all of the data are taken into account. Therefore, it may end up being unrepresentative of the data set as a whole.

Answer 99

It is the most sensitive measure of central tendency as it takes into account the exact distance between all of the values of the data. This means it is more representative of the data as a whole. - The sensitivity of this measure of central tendency can be distorted by one (or a few) extreme values and thus end up being misrepresentative of the data as a whole.

Answer 100

- normal distribution - psoitvely skewed disrubution - negatively skewed distribution

Answer 101

A normal distribution has a number of defining features: · The mean, median and mode are all in the exact mid-point · The distribution is symmetrical around the mid-point · The dispersion of scores or measurements either side of the mid-point is consistent and can be expressed in standard deviations.

Answer 102

This is when most of the scores are bunched towards the left.The fact that there are a few high scores has a strong effect on the mean, which is always higher than the median and mode in a positive skew.

Answer 103

This is when most of the scores are bunched towards the right. The mode is to the right of the mean because the mean is affected by the extreme scores tailing off to the left.

Answer 104

p<0.05 Probability less than 5% i.e. there is a less than 5% probability that the difference/effect/relationship/association that has been found is due to chance. Therefore we can conclude the result is significant p<0.01 - human cost or if the study is a one off that can't ever be repeated

Answer 105

All of your results get put into one mathematical formula called a statistical test. Out of this comes one number that we call our calculated value. This is then compared to one number in a big table of numbers called a critical values table. The one number is called a critical value. Comparing these two allows you to work out if the result is significant or due to chance.

Answer 106

- The researchers are looking for a difference betwen their conditions - A related design (repeated measures or matched pair design) has been use - The level of measurement in the study is nominal data - lookign for DV

Answer 107

1. Identify your three categories. 2. Calculate the number of participants in each category. 3. Assign the category where there is no difference a 0 sign. These participants are then removed from the rest of the test. 4. Assign one of the other two categories a + sign and the other a – sign. 5. Out of the + and – sign categories, identify the category with the smaller number. This number is S.

Answer 108

The calculated value of ______ is greater than/smaller than/equal to the critical value of _______ (p<________, _____ - tailed test, N =________). This means that the result is/is not significant. This means that we can accept/reject the null hypothesis that ________________________________. [If your result is significant, you then add] This means that we can accept the alternative hypothesis that _________________________________________________________. However, because the significance level was __________, there is still a _______________ probability that the results would have occurred even if _____________________.

Answer 109

1. Level of measurement 2. Is the study a test of difference, correlation or association? 3. Experimental design

Answer 110

A related design is a when a matched pairs or repeated measures experimental design has been used. (if for correlation say related data)

Answer 111

when an independent groups experimental design has been used. (if for correlation say unrelated data)

Answer 112

if not, difference is generally for experiments - they’re looking for a change in the DV across conditions. Correlations are relationships where both variables are ordinal or interval data. Associations are relationships where both variables are nominal.

Answer 113

unrelated design - chi-squared related design - sign test cirrelation/association - chi-squared

Answer 114

unrelated - Mann-Whitney related - Wilcoxon Scorrelation/association - pearman’s rho (correlation so related design)

Answer 115

Unrelated t-test Related t-test correlation/assosication - Pearson’s r (correlation so related design)

Answer 116

Can Simon Cowell make winners sing under real pressure nominal -> unrelated, related, correlaiton/association oridnal -> unrelated, related, correlaiton/association interval -> unrelated, related, correlaiton/association

Answer 117

1. Dependent variable 2. Level of measurement 3. Test of difference, correlation or association? 4. Related or unrelated design? (or data is related/unrelated) 5. Appropriate statistical test

Answer 118

* Which significance level was used * Whether a directional (one-tailed) or non-directional (two-tailed) hypothesis was used * How many participants were studied or the degrees of freedom in the study

Answer 119

* The calculated value of ______ is greater than/smaller than/equal to the critical value of _______ (p<________, _____ - tailed test, N =________). This means that the result is/is not significant. This means that we can accept/reject the null hypothesis that ________________________________. [If your result is significant, you then add] This means that we can accept the alternative hypothesis that _________________________________________________________. However, because the significance level was __________, there is still a _______________ probability that the results would have occurred even if _____________________.

Answer 120

this is when the null hypothesis is rejected and the alternative hypothesis is accepted when the null hypothesis is ‘true’ i.e. saying a result is significant when it is due to chance. This is often called the error of optimists. This often occurs when the significance level is too lenient (p≤0.10). The likelihood of making a type 1 error is always the same as the significance level.

Answer 121

Type 2 error: this is when the null hypothesis is accepted but should have been rejected because the alternative hypothesis is ‘true.’ This is often called the error of pessimists. This often occurs when the significance level is too stringent (p≤0.01)

Answer 122

1. Keep the tailed test and the N/df the same e.g. still a two-tailed test, still N=10 2. Change the significance level to the smallest one you can --> p<0.02. In exam questions, go as far to the right in the table as you can, and use the smallest significance level possible. 3. Check if the calculated value is still significant or not --> it's not 4. If it is, you can be confident you haven't made a type 1 error. If it's not, you likely have made a type 1 error. --> we have made a type 1 error

Answer 123

Any method in which a person is asked to state or explain their own feelings, opinions, behaviours and/or experiences related to a given topic

Answer 124

A set of written questions (items) used to assess a person's thoughts and/or feelings

Answer 125

A 'live' encounter (face-to-face or on the phone) where one person (interviewer ) asks a set of questions to assess and interviewee's thoughts and/or experiences, questions may be pre-set (structured interview) or may develop as the interviews progresses (unstructured interview)

Answer 126

Interviews made up of pre-determined set of questions that are asked in a fixed order

Answer 127

Interviews made up of no set questions, general aim that a certain topic will be discussed, and interaction tends to be free-flowing, interviewee encouraged to expand on their answers

Answer 128

An interview that uses techniques from both structured and unstructured interviews - many interviews are likely to be semi-structured

Answer 129

Answers in interviews may be influenced and distorted in some way by the presence, appearance or behaviour of the interviewer yep – it's researcher bias just renamed when it occurs in an interview

Answer 130

Questions for which there is no fixed choice of response and respondents can answers in any way they wish - tend to produce qualatative data

Answer 131

Questions for which there is a fixed choice of response determined by the question eg. yes/no, 1-10 - tend to produce quantitative data quantititive

Answer 132

- Give a fixed number of responses - Easy to analyse and compare since it is quantative - Can convert qualitative closed questions into quantative - can make sure that specific questions that are pertinent to the topic of research are answered because they can restrict the options

Answer 133

- May lack depth and detail in answers compared to open questions - The answer options may not represent what people actually think, feel or did

Answer 134

- Doesn't have fixed range of answers so respondents are free to answer in any way they wish - Contain wide range of different responses

Answer 135

- Hard to analyse the responses - Hard to compare the responses since they're opinions which are subjective so makes it harder to categorise

Answer 136

- Make the questions understandable for participants - don't use technical language that only professionals understand - Write the questions without any bias and use neutral langauge - Don't lead the question to a specific answer - Don't use double negatives as this may confuse the reader - Don't use double-barrelled questions (two questions in 1) as the reader may agree with one of them but not the other

Answer 137

- Make sure all questions are standardised - Conduct the interview in a quite place away from other people - Begin with neutral questions to make the interviewee more comfortable - Remind them about the right to withdraw and their rights - Neutral body language as to not lead the interviewee into a specific direction when answering - demand characteristsics

Answer 138

- Cost-effective- can gather large amounts of data quickly because they can be distributed to large numbers of people - Can be completed without a researcher present, reducing effort involved - Data that questionaires produce is often straight forward to analyse particularly if questionnaire comprises of close questions - quantitative data - Can visually represent data using statistical analysis, graphs and charts - more anonymous as not watched by the researcher while completing it so may be more honest, yielding more valid data

Answer 139

- Responses given may not always be truthful - social desirability bias - Often produce a response bias, where respondents tend to reply in a similar way, may be because respondents complete the questionnaire too quickly and fail to read questions properly - more difficult to ask questions if participants don't understand what they're being asked as the researcher isn't there, therefore they may interpret the question in a way that wasn't intended or may guess, leading to less valid data - it may be harder to express all of their thoughts in writing and so they summarise - leading to less valid data

Answer 140

- Straightforward to replicate due to standardised format - Reduces differences between interviewers

Answer 141

- Doesn't allow interviews to deviate from topic or explain their questions - limits richness from data collected and unexpected information - Little opportunity to build a trust between interviewer and interviewee - can't deviate from the set questions so unexpected answers can't be pursued in more detail. - they feel more formal because of the set questions and so participants may feel uncomfortable, leading them to share less information and so less valid results

Answer 142

- More flexibility in an unstructured interview than structured, interviewer can follow up points as they arise, more likely to gain insight into the worldview of the interviewee including unexpected information Gain more trust from interviewee? Yes quite possibly from it being more conversational

Answer 143

- May lead to increased interviewer bias - Analysis of data is not straightforward, may have to look through irrelevant information and drawing firm conclusions may be difficult - Risk that interviewees may lie for reasons of social desirability - it's difficult to analyse the data because the individual responses are so different, this make it hard to summarise the results - esearcher bias may be an issue as the researcher will be developing questions on the spot which may unintentionally influence the outcome. They also are involved in the interpretation of the data where they could again unintentionally influence the outcome (less valid results)

Answer 144

to determine the likelihood (probability) that the effect/difference/relationship/association found is due to chance

Answer 145

1. Significant - there is a difference/relationship/association between the two variables i.e. the alternative hypothesis is correct - we can't ever be 100% certain about this one because of potential extraneous variables and we haven't studied every person in the target population 2. Chance - the results would have occurred even if there is no difference/relationship/association between the variables i.e. due to something other than the IV, fluke, coincidence i.e. null hypothesis is correct can be confident in this one

research methods Flashcards

(179 cards)