Research Methods Flashcards

Question

Outline and evaluate the different measures of tendency

Answer 1

Mean: Most representative (accounts for every person); Use when no anomalies present (lower score and brings mean/average down = causes mean to be skewed) - Strength: Makes use of all the values = Most representative - Weakness: Easily disorted by extreme values Median: Better reflection of data than mean - Strength: Easy to calculate (unaffected by extreme values) - Weakness: Ignores most values = isn't as representative as mean Mode: Only important for distributions, otherwise never used - Strength: Unaffected by extreme values - Of little use when there are no repeat values or is bi-modal (multiple modes)

Answer 2

Range: - Strength: Easy to calculate - Weakness: Doesn't tell us about spread of values and is affected by extreme values Standard deviation: - Small standard deviation (low value) = Data is consistent; all participants reacted similarly (scores close) - Large standard deviation (high value) = Data is dispersed; participants affected differently (scores widely spread) - Strength: Includes all values = more precise measure of dispersion; can tell how each individual participant performed - Weakness: Difficult to calculate and can be skewed by anomalies

Answer 3

Normal distribution: Spread of frequency data which forms a bell curve (mean, median and mode all located at the highest peak) Skewed distribution: Spread of frequency data that clusters to one end - Positive skew: Long tail on the positive (right) side = most distribution - Negative skew: Long tail on the negative (left) side = most distribution

Answer 4

Bar graphs: Used when data is divided into categories (discrete data) - Frequency of each variable is represented by height of the bars - x-axis = different categories (e.g. eye colour - each bar represents a different eye colour) - y-axis = frequency (e.g. how many people have the colour eyes) Histographs: Continous data rather than discrete (bar chart/graph) - x-axis = equal sized intervals of a single category (e.g. age - an example of continous data) - y-axis = frequency within each interval (e.g. number of children) Line graphs: Continous data; use points connected by lines to shown how something CHANGES in value (e.g. overtime) - x-axis = continous interval e.g. day (Mon, Tue, Wed, Thur, Fri, etc) - y-axis = frequency e.g. number of hotdogs sold (per day) Scattergrams: Depict associations rather than differences between co-variables (correlations); represents the strength and diection of a relationship between co-variables in a correlational analysis - x-axis = e.g. exam score - y-axis = e.g. number of hours spent revising (i.e. is there a relationship between the score achieved and the number of hours spent revising? Positive/negative/zero correlation?)

Answer 5

Peer review is part of the verification process which decides whether research is scientifcally acceptable or not. It consists of a system used by scientists to determine whether research findings can be published in scientific journals Aims of peer review: - To allocate funding (decide whether or not funding for a proposed research project should be awarded) - To validate the quality and relevance (re-run calculations for conclusions made) - To suggest ammendments or improvements - Evaluate ethical treatment of participants

Answer 6

Finding an expert: - Experts should be objective and unknown to the researcher - Experts in the same particular field may try to sabotage because they need funding (in direct competetion for limited research fundings and do not want other people's research to get funding instead) Anonymity: - Peer (expert) remains anonymous for more honest appraisal - A minority of peers use anonymity to critise rival researchers because they are in direct competetion for limited research fundings, so in some journals where open review has taken place the names of the reviewer(s) are public Publication bias: - Editors want to publish significant "headline grabbing" findings and more positive results to increase credibility and circulation of publication - Therefore some research may be ignored (if it does not meet this criteria) which creates a false impression for psychology as journal editors are being selective in what they publish Buying ground-breaking research: - Peer review may suppress opposition to mainstream theories to maintain the status quo with particular scientific fields - Reviewers are more favourable to research which matches/supports their own - Established scientists are more lkely to be chosen by prestigious publishers so findings that support the current opinion will be passed whereas new and innovative research is ignored as it has not been peer reviewed because it challenges the established order - Therefore peer review slows down the rate of change

Answer 7

If psychologists understand why we behave and think the way we do, they can create methods which positively contribute to the population and the economy A third of all absences from work are due to mental health, including depression, anxiety and stress Research on drug therapies means: - Less absence from work - Less pressure on NHS - Less pressure on emergency services - People will be able to go into work = earn a better salary = better quality of life Research into the accuracy of eye-witness testimony means: - Less money spent on investigation - Less people wrongfully convicted (who would thus be more likely to sue) - Less crime as the right people are arrested

Answer 8

The sign test is a statisical test to determine whether results are significant or not - We use probability to test is the results occurred by chance or not - Accepted level of probability = p<0.05, meaning there is less than 5% probability that the results occurred by chance (IV was responsible for at least 95% of the effect measured in DV) It is only appropriate to use the sign test when: - It is a test of difference (an experiment) - There is related design (repeated measures/matched pairs) - There is paired (nominal) data - each participant is categorised as either having a plus or minus sign

Answer 9

1.) Convert data to nominal data (paired) by finding out which particiapnts produced a high score in the 2nd column and which produced a lower score in the 2nd column. Make not of this with a + or - (if the first score is higher then +, if first score is lower then -, if scores stay the same then =) 2.) Add up the + and - signs respectively to get two totals. 3.) Work out your S-value

Research Methods Flashcards

Students should demonstrate knowledge and understanding of the following research methods, scientific processes and techniques of data handling and analysis, be familiar with their use and be aware of their strengths and limitations: