all research methods

a broad statement on what will be tested

a prediction that is precise and can be tested

predicts the relationship between the two variables

students perform better after they have eaten compared to when they are hungry

has to be written in a measurable and testable form. it can't be ambiguous

Students who take a 20 mins maths test 30 mins after lunch will get more marks out of 20 than students completing the test 30 mins before lunch

the thing the researcher manipulates or changes

this is the thing the researcher measures

a small study carried out to test the feasibility of a larger one

they are not a participant. they are playing a role

check the tasks aren't too easy

alternative, null, directional, non-directional

prediction of what will happen to the DV when the IV changes and any observed differences will be due to chance

states that the IV will have no effect on the DV and any observed differences will be due to chance

predicts that there will be a difference between two conditions or groups of participants without stating which one is better

states which group is better of the predicted differences between the two groups

chosen when there is no previous research or lots of controversy

when previous research implies a pattern of findings

variables which must be eliminated or controlled otherwise they may affect the dependent variable

these are variables to do with the situation and affect the behaviour of participants

time of day, lighting

variables to do with the participants

age, gender, social class, ethnicity

extraneous variables which have not be controlled and may affect the results

an aspect of the research situation which leads to the participant to guess the aim and change their behaviour accordingly

deception and the single blind method

anything the researcher does which can affect how the participant behaves and the subsequent data collected

double-blind method

this is when participants are not told which condition they are in

neither the researcher or the participant know about the hypothesis or which condition they are in

randomly allocating participants to groups so any difference cancel eachother out with a large sample

using standardised procedures and instructions so participants have the same experience

consistency of measurement

different researchers conducting the same study on the same person

two observers or more records the behaviour and check that they have it similar

the extent to which a test measures or predicts what it is supposed to

it is certain that the IV has cause the DV to chnage

the extent to which results can be generalised to other settings

the extent to which an experiments findings can be generalised to real life

the extent to which research looks as though it is doing what it is supposed to on the surface level

the extent to which findings can be generalised from the sample to other populations

the extent to which a new measure compares to a previous validated measure

the extent to which research findings hold true over time

The extent to which findings can be generalised from the sample to other populations

test retest

face validity- eyeball test

participants are tested once and then tested again at a later date

the group of people the researchers want to apply their results to and who they draw their sample from

a small number of people from the target population who take part in the investigation

this may occur if the sample selected is not represented of the target population

by making the sample as large as possible

using people from the target population available and willying to take part

participants volunteer to take part in the study

chooses participants in a logical way from the target population

subgroups within the target population are identified and participants are obtained from each subgroup in proportion randomly

a sample of participants produced by using a random technique and everyone in the target population has an equal chance of being chosen

representative as everyone has an equal chance of being chosen

convenient as coasts less time and money

convenient as it requires less input from the researcher

representative as sub groups in your sample are proportional to the target group

it is convenient if the sample is small as you get a complete sampling frame used

may be unrepresentative for small samples as participants with similar demographics may be selected

can be biased as sample only drawn from one area and may not reflect the wider target population

unrepresentative as volunteers may be a certain profile of people eg, helpful so it may be harder to generalise to the target population

requires lots of time to figure out the proportions for large samples and recruit the right number of people for each sub group

can be difficult to obtain a complete sampling frame for large populations so may not representative of the wider population

numerical data

descriptive data

in experimental based research methods

in case studies, unstructured observations and interviews

they describe and summarise quantitive data

measures of central tendency and measures of dispersion

mean, median, mode

the statistical average of the data set

the middle number of the data set

most frequent value in the data set

uses all values so most sensitive measure

can be distorted by anomalies

unaffected by extreme values

only takes into account 1 or 2 scores

unaffected by extreme values

not useful in small data sets

range and standard deviation

it is the difference between the largest and smallest number of a data set

measure of the distribution on scores. the larger the standard deviation the larger the distribution of scores

quick and easy to work out

it takes into account all scores so is a sensitive measure

it can be distorted by less meaningful data

less meaningful is data is not normally distributed

change/original x 100

the median, mode and mean all occur in the same place at the peak of the curve, and is symmetrical

more high scores than low scores- ceiling effect

more low scores than high scores- floor effect

it is concerned with what is deemed acceptable

can do can't do with participants

consent, deception, confidentiality, debriefing, withdrawal, protection from harm

gathering primary data by observing relevant people, actions, and situations