Methodology (social)

Hypotheses

Hypothesis: a specific testable prediction about what you expect to find  after analysing the data from your participants. (aka, an alternative hypothesis)Null Hypothesis: states there is no such effect except that found by chance

Questionnaires

A self-report method that involves the systematic gathering of data from large numbers of people, usually by carrying out a questionnaire. Carried out on samples. Can be conducted face to face, by post, by telephone or via the internet.

Open-ended questions

Are those in which the researcher does not restrict the range of available answers - infinite range of possible answers.Produces Qualitative data - rich in detail or description, usually in textual or narrative form.Strengths:Lots of detail/ more depth (allows individual differences)Can't bias the question (lowers experimental bias)Weaknesses:Hard to interpret/ understandDifficult to compare answers/ analyse

Closed Questions

Answers are restricted. The researcher determines the range of possible answers. Best used when factual information is required. E.g. Likert or ranking.Generates quantitative data - Numerical data, measurements of quantity or amount. Can be easily measured.Likert scale - a closed question that uses a restricted number of answers to get usually quantitative data.Strengths:Easy to quantify and analyseCan get a majority answer/ more valid dataWeaknesses:May lack realism due to forced choiceMay be biased due to how the question is phrased (leading questions)

Evaluation of Questionnaires

Strengths: can be large scale replicable so high reliability lower experimental bias than interviews

Weaknesses: rigid so not much room to follow up on questions problems that come from self-report method (see later slide)

Interviews

Qualitative data is generated from interviewsStructured: pre-set order of questionsSemi-structured: there is a schedule of questions that should be answered but the researcher has freedom to follow up on some responsesUnstructured: questions are open and structure is flexible

Evaluation of Interviews

Strengths: In-depth data Can be more flexible than questionnaires

Weaknesses: Likely to be small-scale Can't be replicated - low reliability More likely to get experimenter bias than questionnaires Problems that come from self-report methods (see next slide)

Weaknesses of self-report methods

Social Desirability: when respondents say what they think they ought to say - reduces internal validity because ppts don't give an accurate or true responseDemand Characteristics: guessing the purpose of the study - not acting in a realistic way, not giving accurate responses which reduces internal validityResponse Bias: the response does not reflect their true feelings - the answers will be worthless as the ppt won't fill it in properly - reduces internal validity

Reliability

Reliability: consistency of the dataInternal Reliability: refers to the consistency of a tool, or a procedure, also refers to how consistent each individual is on self-reportingExternal Reliability: the degree to which results will be consistent over trials. Same results after repetition

Ways to increase reliability

Test-retest: checks consistency. Given to same people on 2 occasions. Strong relationship between 2 results indicates good reliability.Split Half: compares the results of one half of a test with the results from the other half (where Q's basically ask the same thing) - similar results suggest good internal reliabilityInter-observer/ Inter-rater: use 2 or more interviewers (only used in interviews) - similar results indicates good reliability

Validity

Validity: extent to which the IV is known to cause the DVFace validity: does it measure what it set out to?Content validity: is it appropriate to investigate what we want to find out? Does it cover all aspects?Concurrent validity: test against a known/ current questionnaire. Similar results shows concurrent validity.Predictive validity: should predict/ reflect real-life behaviour

b9d14224-441c-4506-ab03-70b3952dcc26 (image/jpeg)

Pilot Survey

Sampling Methods

Sample: a collection of people taken from the target population which accurately reflects the target population - drawn from the sampling frame.Target Population: the whole group of people about whom research intends to establish a fact.When a sample reflects the whole target population, it's described as being representative. A more representative sample increases population validity.

Random Sampling

When a random method is used to select a sample from a list of the whole target population.Strengths: Quick, cheap and easy No experimenter bias Sample can be checked mathematically Weaknesses: Individuals may not be suitable Hard to achieve true randomness Not all people picked may want to take part

Volunteer/ self-selected

When the research project is advertised with a poster, newspaper ad etc and interested people approach the experimenter to become ppts.Strengths: Right type of people found, can pick and choose to a degree Ethical - they volunteered More likely to get full cooperation Weaknesses: Takes time for people to volunteer and get a large sample Volunteer bias May only get certain types of people

Stratified

Where ppts are identified from a list of the whole target population, but experimenters seek to acquire a precise representation of the population.Strengths: good sample found with right characteristics - high population validity not biased by the experimenter Weaknesses: takes a lot of time and costs more difficult to get a full breakdown of a population

Opportunity

Makes use of sample already gathered for some other reason/ asking people that are already there.Strengths: quick, cheap and easy method (simpler) efficient ethical (can give consent there and then) Weaknesses: unrepresentative sample More chance of bias towards people that are easier to access

Quantitative Data Analysis

Includes measures of central tendency (mean, mode and median) and measures of dispersion (range and standard deviation)

Mean

Add up all the numbers and divide by the number of numbers.Advantages:Makes use of all the numbers in a data set and is therefore a sensitive measureDisadvantages:Affected by extreme scores and can misrepresent the numbers as a resultCan only be used with certain types of data

Mode

The most common number, calculated by putting similar scores together and counting which one occurs the most frequently.Advantages:Can be used with nominal dataProvides information about frequencyDisadvantages:The data may have 2 modes (bimodal)

Median

Central value, arrange the scores in order and find the middle valueAdvantages:Not affected by extreme scoresCan be used with ordinal dataDisadvantages:Not as sensitive as the mean, as not all values are reflected

Range

The measure of the spread of a set of scores, shown by the difference between the highest value and the lowest.Advantages:Easy to calculate and shows extreme valuesDisadvantages:Affected by extreme values and doesn't take into account the number of observations in the data set.Doesn't give info on whether the scores are clustered around the mean or spread out.

6430604f-1fe8-4bba-b3b9-664ad34fdcfe.gif (image/gif)

Standard Deviation

The measure of the spread of data around the mean. The higher the value the more variation in your scores. Bigger spread = bigger graph = bigger standard deviation. More consistent scores = smaller/tighter graph.Advantages:Gives a more precise and sensitive measure of dispersion than the range, as all values are taken into account.Disadvantages:More complicated to calculate than the range.May hide some characteristics of the data.

Thematic Analysis

A systematic research technique for analysing transcripts of interviews, documents, text, including adverts, children's books, TV programmes, cartoons, films, song lyrics, newspapers, magazines and websites.Used with qualitative data and aims to turn qualitative data into quantitative data which can then be analysed using descriptive and inferential statistics.Steps: A coding system is created A pilot study is often conducted Conduct the analysis Turn the data into quantitative displays (e.g. bar charts) Reliability might be checked

Thematic Analysis Evaluation

Strengths: great to study emotions and motivation can establish what caused the behaviour can study rare behaviours in detail

Weaknesses: not scientific (coding and analysis is subjective) can't generalise the situation being observed how on person views the data may be different from someone else (low reliability) language may be ambiguous and misinterpreted (low validity)

Ethical Issues

A difficulty, tension or problem which arises during the course of the research processThis may be because there's conflict between what the researcher wants to study (aim) and how he or she goes about it, with potential negative consequences for ppts or those being represented for.DR C DRIP

Lack of Debriefing

Standard practice to debrief ppts in full at the end of a study.Should restore ppts to the state/condition they were in prior to the study, all deception should be cleared up.Not always possible in observational/ covert studies

Lack of right to withdraw

Ppts are given the right to withdraw themselves from the study at any time, and the right to withdraw their data too.If payment has been agreed, and a ppt withdraws, payment must still be made.Ppts must not be coerced into continuing with a study (Milgram)Ppts must suffer no negative effects when they leave.

Lack of confidentiality and privacy

Anonymity: data is not identifiable to them by name.Confidentiality: relates to how the data is stored and who sees it. (only people with a need to know/see the data should see it)Privacy: not listening into what a ppt is saying, or looking at what they're writing and doing.Ppts should be observed in places where they would expect to be observed by strangers, and think that they're likely to be.Rights should be explained to ppts via standardised instructions.

Use of deception

Deception: when you lead to someone to believe something that isn't true. (should use minimal deception)Small amount of passive deception is often necessary when describing the aims of a study - otherwise can lead to demand characteristicsCan be active (where an actual lie is told) or passive (where something is omitted)

Lack of informed consent

Researchers explain what the aim of the study is and what they'll be asked to do. Ask the ppt if they're happy to take part. Consent is a particular issue in observational studies as none is given.Where deception is involved, consent is no longer informed.Consent is needed from parents for ppts under 16.For vulnerable ppts, consent is to be obtained from an appointed adult or guardian.

Lack of protection from harm

The risk of harm should be no greater than ppts would experience in their everyday life.This includes physical and psychological harm - including embarrassment and loss of self-esteem.Ppts (human or animal) should be subjected to minimal harm.

Giving advice

Researchers aren't supposed to advise ppts - should refer them to a professional who can help them address physical or psychological problems.