aos 4: research methods

research question

based on the topic that can be investigated and tested.

aim

statement regarding the topic that is being investigated or tested. always starts with ‘to investigate’

variable

any factor that can change in amount or type over time

controlled variable

must be held constant to remove potential effects on the dependent variables.

dependent variable

variable we measure

independent variable

variable that is manipulated, changed or varied in some way to measure the effect on the dv

population sampling

how the sample is chosen from the population

beneficence

where potential research outcomes are maximised and risks of harm are minimised.

confidentiality

\
right to privacy, participants must not be identified in any way in the results

voluntary participation

participants have the right to refuse to take part in the study, no pressure or deception to ensure participants take part

withdrawal rights

participants have the ability to leave the study at any stage regardless of possible effects on the results. they also have the ability to remove their results at the conclusion of the study.

informed consent

participants must be told about the true nature of the study, their participant rights and any potential risks. they must give written agreement to take part. minors or those who aren’t of sound mind must have written agreement by their guardians

debriefing

takes place at the conclusion of the experiment, and helps ensure that the participant understands the aim, results and conclusions. support/counselling is also provided to ensure no lasting harm has been created. debriefing is especially essential with studies involving deception

random sampling

every member of the population has an equal chance of being selected to participate.

hypothesis formula

POP, 1/2 IV , D, DV, 1/2IV

population

entire group of people researchers are interested in

sample

a smaller group of people taken from the population that participate in th study

stratified sampling

ensures each subgroup of a characteristic is represented in the sample in the same proportions as it is in the population

justice

ensure a fair distribution of benefits and burdens with the population of research interest

ethics

protect the rights and welfare of participants, and ensure that the research study is designed in ethically appropriate ways in accordance to the ethical guidelines

what are the 5 ethics guidelines

beneficence, respect of persons, integrity, non-maleficence, justice.

respect for persons

properly regard welfare, rights, beliefs and customs of individuals involved in study

integrity

researcher’s responsibility to conduct research and report results in an honest ethical manner, whether it reported any literature reviews undertaken prior to the study

non-malificence

avoiding causing harm. harm shouldn’t be disproportianate to the benefits from any position of action

what are the 6 participant rights

voluntary participation, informed consent, confidentiality, deception, withdrawal rights, debriefing

deception

when participants are not told of the true nature of the study.

when is deception warranted?

if it is approved by the ethics committee, so that the value of research outweighs the need for deception and appropriate debriefing and counselling procedures are in place to ensure that there is no lasting psychological or physiological harm to participants.

between subjects design

participats seperated into different groups and only do the experiment once, therefore only receiving one level of the iv.

disadvantage of between subjects design

individual participant differences

advantage of between subjects design

no order effects to control

within subjects design

participant takes part in all conditions, the experimental and control groups.

advantage of within subject design

minimises the impact of individual participant differences

disadvantage of within subjects design.

creates order effects, where participants perform better in the second conditioned because they have practiced.

how to avoid order effects?

using counterbalancing where each group of participants receives treatment in a different order.

mixed design

includes elements of both between and within subjects design.

primary data

collected directly by the experimenter

secondary data

collected from another source

quantitative data

quantities represented in numbers

qualitative data

non-numerical descriptions, concers or concepts

subjective data

individual or personal experiences or opinions of participants. collected through self-report and cannot be directly observed.

objective data

collected through direct observation or measurement. should obtain same results regardless of experimenter.

case studies

in depth investigation of some behaviour or event of interest

case studies: 4 advantages

good when a large number of participants aren’t available, avoid artificiality, can be conducted over a prolonged period, valuable source of hypothese

case studies: 4 limitations

cannot be replicated, cannot be used to test hypotheses, time consuming, cannot generalise

field work

inquiry or investigation of an issue involving observation and interaction with a selected environment to determine correlation

observational studies

collection of data through watching and recording clearly visible behaviour. can be structured, unstructured or semi-structured.

observational studies : 3 advantages

not influenced by artificiality, behaviours cannot be reproduced, doesn’t require participant cooperation

observational studies : 3 limitations

observer bias, privacy violation, difficult to determine causes of behaviour

yarning circles

sharing of stories, understandings and opinions. all povs are equally valid.

simulation studies

reproducing situations of research interest in a realistic way to investigate behaviour or mental processes. used if a researcher cannot access a particular setting

simulation studies : 2 advantages

no danger or harm, easily replicatable

simulation studies : 2 limitations

environment is artifical, cannot generalise

self-reports

written or spoken responses based on emotions, feelings and behaviours when conducting research. include questionnaires, rating scales and interviews

questionnaires

written set of questions to draw out self-report info

4 advantages : questionnaires

efficient, anonymous, consistent, easily summarised data

4 limitations : questionnaires

innacurate answers, influenced by social desirability, too structured, sujective

interviews

questions asked by researchers

2 advantages : interviews

detailed info, different level of structure

4 limitations : interviews

time consuming, social desirability, subjective, hard with large samples

rating scales

person selects a number to reflect quality of a question

2 advantages : rating scales

easy and quick, collected at once

2 limitations: rating scales

subjective, don’t give reasons

correlational studies

non-experimental study which researchers, observes and measures the relationship between two or more variables.

2 advantages : correlational studies

used to test hypotheses, indicate patterns or trends

2 limitations : correlational studies

can’t draw firm conclusions, can’t control unwanted variables

random errors

due to chance, factor or variation in a measurement. unknown or uncontrolled factors affect measurement process. affect precision.

systematic errors

produced by a factor that is consistently favouring one condition. associated with a flaw in the research design. affect the accuracy of results

personal errors

fault or mistake made by researcher

uncertainty

lack of exact knowledge of the value of the quantity being measured

accuracy

how close measurements are to the true value of the quantity being measured

precision

how close measurements are to each other.

extraneous variable

any other variable other than the iv that may cause an effect on the dv. when left uncontrollled they become confounding variables

confounding variable

unwanted variable that has a systematic effect on the dv. mean no conclusion can be made from experiment

addressing individual participant differences

changing experimental design or random allocation

addressing order effects

counterbalancing

demand characteristics

when a participant’s response is changed by their belief they should respond in a particular way

placebo effect

participant’s behaviour changed based on belief they are receiving experimental treatment

addressing demand characteristics

single blind. participants don’t know if they are in the cr or er.

addressing placebo effect

single blind participants don’t know if they are in the cr or er.

experimenter effect

the experimenter’s personal characteristics, actions or treatments of the data affect the dv

addressing experimenter effect

double blind, a third party source allocates cr and er, so experimenter doesn’t know

situational variables

when instructions are given to each participants or groups differently

addressing situational variables

standardised intructions and procedures

small sample size

with a small sample size, there is not enough participant variability present. making it unlikely that the sample is a true representation of the population

addressing small sample size

increase sample size

mean

average of data

median

middle number in data set

mode

most frequently occuring number in data set.

standard deviation

how far scores within a set spread out or deviate. low sd means there is little variation and causes clustering around the mean. high sd means a greater variation in data sets.

outlier

extreme measurement which significantly differs from others in data set

graph conventions

dv = vertical axis, iv = horizontal axis

validity

accuracy. the extent which the research study measured what it intended to measure. these should match the requirements of the investigation

internal validity

extent which a study investigated what it claimed to investigate

external vailidity

extent which results obtained can be applied beyond the sample that generated them.

reliability

consistency. extent at which results are consitent, dependable and stable.

repeatability

a study is repeatable if replicated under the same conditions and produces similar results

reproducibility

a study is reproducible if replicated under different conditons and produces similar results

conclusions

the final decision about the effect of the iv on the dv. mention whether hypothesis was supported, direction of effect and data.

classification and identification

classification is the arrangement of phenomena, objects or events into manageable sets, whereas identification is a process of recognition of phenomena as belonging to particular sets or possibly being part of a new or unique sets.

literature review

the collation and analysis of secondary data related to other people’s scientific findings to answer a question. can also be used as preparation for an investigation to generate primary data. useful to help formulate hypotheses and methodlogy for an experiment