Research Methods Flashcards

Question

Reliability of measurement can be improved by a pilot study. How?

Answer 1

- Materials should be tested to ensure that they yield reliable results from participants i.e. by measuring test re-test and split-half reliability. This should aim to ensure question are not affected by irrelevant factors. - Researchers should be trained to ensure that they are reliable over time for an individual ( to ensure intra-rater reliability) and with each other across a team to ensure inter-rater/observer reliability. This should eliminate inconsistencies in ratings and observations.

Answer 2

1) Population refers to those who the study is about and who the results apply to 2) ... if in the sample, you have people who are typical of the population. 3) To generalise is to claim that what is true of the sample is also true of the population 3) Random- involves the use of a sampling population/frame i.e. a list of possible particpants from whom the actual sample is chosen via some unbiased mechanism e.g. a computer programme so that all members of the sample have an equal chance of being picked. - Opportunity sampling - involves using whoever is available for the research e.g. patients in therapy, students on a course. - Volunteer - involves asking volunteers by publicizing the research where the target audience will see it

Answer 3

1) a) Deliberate or unwitting bias by the researcher in selecting particpants - a researcher may choose a ppts who he believes will give him results that support his hypothesis. b) The use of unrepresentative sampling populations from which samples are taken - If just on one country, it might not be representative of other cultures. c) Chance, or random error, when taking samples, which is more likely with small samples. - when taking random samples it is possible to get an unrepresentative sample just by chance. This can happen in any random sample though it if more likely to occur if a small sample is being chosen. This can be addressed by the use of stratified sampling/ 2) Stratified sampling - divide the sampling frame into groups to be represented in the sample. Sampe randomly from within these groups. Then if a sampling frame is unavailable,.... ..... draw up quotas to be represented, then recruit volunteers till quotas are full. 3) Stratified sampling eliminates the possibiliy of the very unrepresentative samples that can be produced by random sampling. HOWEVER they are expensive to run - - Quota smapling also improves the likelihood that the sample is representative by ensuring that the proportions of each strata in the sample are representative. HOWEVER, there are problems with using volunteers - you will get a certain e.g. in a studies of the impact of attachment few very bad parents volunteer leading to a bias sample.

Answer 4

1) Informed consent - participants should give informed consent. i.e. be made aware of anything that might affect their willingness to particpate. This should include an information sheet about the study, how ethical issues will be dealt with and a declaration of consent. - ---The main aim is to acquire informed consent, i.e., participants are aware of anything that might affect their willingness to participate a) Information sheet about the study * Statement of purpose of study, insofar as this needs to be revealed * What participants will be asked to do, i.e, procedures of the study, in so far as these need to be revealed * This needs to be carefully considered, ensuring that not too much, too little or misleading information is given b) How ethical issues will be dealt with * How anonymity and confidentiality will be ensured * Assurance about participant’s right to withdraw c) Declaration of consent * Include a declaration form for participants to tick and sign, e.g., 2) The right to withdrawal should be clear at the onset of the study i.e. in the consent form. Participants also have the right to withdraw their data retrospectively i.e. after the study. 3) A debriefing should provide participants with any necessary information to complete their understanding of the nature of the research and monitor any unforeseen negative effects or misconceptions. It should include: a) Preliminaries- thanks! * Thank participants for taking part. b) Procedure- completing understanding of the study * Clarify the aim of the study, including anything they need to know to complete their understanding of the study (i.e. anything they wouldn’t have been told when giving consent). c) Ethics- dealing with remaining issues * Ask if they have any questions. * Researchers should ‘identify any unforeseen discomfort, distress or other negative effect of the research’. * Remind participant of their right to withdraw (retrospectively) and to insist that their personal data is destroyed. 4) Children under 16 cannot give their own consent. A parent can give consent on their behalf. In a school setting, children have to give their own consent as well as the head/teacher.

Answer 5

1) Contents 2) Title - clear idea of what the research is about which may include naming the theory of the variables 3) Abstract - a summary of the key points of the research with a list of key terms which is often provided separately. The abstract and list of key terms allow researchers to quickly decide if the research is of interest to them. 5) Introduction to background of the study including key issues and debates and relevant studies. ( the funnel technique should be used so that you start with general material then funnel down to more specific research/theories 6) Hypothesis - a hypothesis should be written before research is conducted so that the researcher does not change it later to match their results if they were different to what they expected 7) Method/procedure - gives a clear account of how the study was conducted - this allows replication of the study and therefore findings can be checked should include - method and technique - design - variables - sampling method - materials - procedure - controls e.g. counterbalancing - ethical issues 8) Results - including descriptive statistics (written account of findings, graphs and tables e.c.t) and inferential statistsics ( details of statistical test). It is crucial to identify a statistical test in advance so that u ensure there is an appropriate stat test that can used 9) Discussion of strengths and weaknesses e.g. reliability, validity and generalisability and possible improvments for future research. They may also include practical applications. 10) References - alphabetical list of sources 11) Appendices includes - consent form - lengthy instruction sheet - original materials - raw data (made anonymous) - debriefing form

Answer 6

A higher standard deviation is showing that the scores are more spread out in this condition. Hence, there was more variation in this condition - e.g. if this condition was for the use of brief CBT, then it implys that the therapy may have been more effective for some than others.

Research Methods Flashcards

(30 cards)