Experiments- research methods Flashcards
Independent Variable
Independent Variable (IV): a feature of the experiment that is manipulated by the researcher in order to observe whether a change occurs in the dependent variable (DV)
Dependent Variable
Dependent Variable: result of that manipulation. The experimental variable that is measured by an experimenter
Level of significance
Level of significance: this indicates the extent to which a set of results are due to chance factors alone rather than the independent variable (I.V)
Laboratory experiments
the IV is manipulated by the researcher and the experiment is carried out in a laboratory or other contrived setting away from the participant’s normal environment
Field experiments
the IV is manipulated by the researcher but this time the experiment is carried out using participants in their normal surroundings
Quasi (or natural) experiments
the IV is naturally occurring (e.g. cloudy conditions versus sunny conditions; morning versus afternoon), not manipulated by the researcher
Advantages of lab experiment
Standardised procedure (same materials for all ps so it’s consistent-reliability)
Control of extraneous variables-validity
disadvantages of lab experiment
High risk of DC(usually)-construct validity
Low ecological validity- doesn’t reflect real life
advantage of a field experiment
Field experiments can offer a more realistic
setting for a study, and therefore can have
more ecological validity
disadvantage of a field experiment
Lack of control can mean it is difficult to
assume that the variable manipulated was actually influencing behaviour and that it wasn’t something else.
Repeated measures Design
this involves using the same people in each condition
Independent Measures Design
this involves using different people in each condition
Matched groups design
this involves using different people in each condition but an attempt is made to make the participants as similar as possible on certain key characteristics (any that might influence the findings). This is done by testing the individuals on the key characteristics, pairing them based on similar scores, and then placing one member of each pair into each group.
Extraneous Variables:
a variable that if they are not controlled they may obscure the effect of the I.V. If this happen then it becomes a confounding variable
An alternative hypothesis
This predicts how one variable (the IV) is likely to affect another variable (the DV). An alternative hypothesis predicts that the IV will affect the DV.
A null hypothesis
This predicts that the IV will not have an effect on the DV. The null hypothesis predicts that any difference seen will be due to chance factors rather than the independent variable.
Two tailed
This predicts that the IV will have a significant effect on the DV (i.e. there will be a significant DIFFERENCE in the results from the different conditions of the experiment), but it does not predict the direction this effect will go in.
One Tailed
This predicts not only that the IV will have a significant effect on the DV but also the direction this effect will go in (i.e. the kind of difference – more or less, quicker or slower – between the conditions of the experiment).
advantages of mean
easy to calculate
presents central point of distribution curve
disadvantage of mean
may not be representative of any scores in the data
affected by extreme scores
advantage of median
Not affected by extreme scores
disadvantage of median
Can be distorted by small samples
advantage of mode
Not influenced by extreme scores
It is the only measure which can deal with non-numerical data (tally frequency on observations). For example the most frequent eye colour in this group.
disadvantage of mode
Not useful if many equal modes
Range
is a measure of how spread
out your data is
Variance
measures how much a set of numbers is spread out. On its own, it gives an idea of how spread scores are from the mean. It comes into use when calculating the standard deviation
Standard Deviation
describes how spread out a group of numbers are. It is a better measure of dispersion than the range because it is less affected by extreme values
advantages of range
Shows us the range of values in the data
disadvantage of range
Can be skewed by outliers in the data
Doesn’t take account of how spread apart the data is – e.g is the data clustered at one extreme?
advantage of strandard deviation
Precise calculation as all data points are included – Less likely to be skewed by outliers
The score is represented in the same unit that the original data is in.
disadvantage of standard deviation
More tricky to calculate than the variance
advantage of variance
A more precise calculation as all data points are included.
Less likely to be skewed by outliers
disadvantage of variance
It is a squared figure – It is not represented in the same unit that the original data is in.
Advantage of Quasi
It allows us to study the effect of the variables psychologist can’t manipulate or change on behaviour.
Disadvantage of Quasi experiments
there is no control over the participants, in terms of social settings how they were brought up, life style, etc. and these may be confounding variables which influence behaviour
Advantage of repeated measures
By comparing each person with themselves the like-hood that individual differences between subjects will confound the study is reduced. This is the best design, therefore for controlling subject variables in an experiment.
This design can use fewer participants than matched participants of independent measures design so may be more cost and time effective.
Disadvantage of repeated measures
The repeated measures design can be affected by order effects such as practice, fatigue and boredom, so it requires counterbalancing to control for these.
If subjects are tested twice or more they may work about the IV ( done by picking up on demand characteristics) and try to behave according to what they believe is expected.
Advantage of independent measures
An independent measures design isn’t affected by order effects as each participants is tested only once in one condition.
This design is also less likely to be affected by demand characteristics than a repeated measures design as each person is tested only once and has less opportunity to work out the hypothesis being tested and act accordingly.
This design is less time communing than matched participants design
Disadvantage of independent measures
An independent measures design does not control extraneous participants variables effectively, so individuals differences between the participants may confound the findings.
Large samples are often needed in order to be sure that any effect of the independent variables is caused by the dependent variable and not by individual differences
Advantage of matched participant design
A matched design isn’t affected by order effects and is less likely to be affected by demand characteristics as each participant is tested only once.
A matched participant design controls participant variables better than an independent measures design as participants are matched on variables important to the study
Disadvantage of matched participant design
The matched participant design isn’t often used as it is very time consuming to match participants.
It is impossible to match participants on enough variables to be sure that there are no possible extraneous individuals differences that might confound the study.
Operationalisation
refers to the process of making variables physically measurable or testable. This is done in Psychology by recording some aspect of observable behaviour that is assumed to be indicative of the variable under consideration. Operationalisation may be required in respect of the IV as well as the DV.
Measures of Central Tendency
express the “average” or “typical” score within a data set. They help us to show the middle point or most frequent number.
Self selecting sampling
This is when people volunteer to take part in the study. Often adverts, posters or leaflets will be distributed which contain details about the research and contact details for participants to use if they wish to take part.
Opportunity sampling
A sample of participants produced by selecting those who are most readily available at a given time and place selected by the researcher.
Random
A technique in which each member of the target population has an equal chance of being selected. For example, all the names of everyone within a given target population could be placed in a hat, and the first 20 to be drawn out could comprise the sample.
Snowball
When participants are asked to contact their friends and family to ask them to also take part in the research; they, in turn, then ask other people.
Strengths of self selecting
- Ethical as participants volunteered
- Relatively easy and particpants are likely to turn up
Weaknesses of self selecting
- Biased as based on volunteers
- possible time and cost of advertising
- may not get any volunteers
Strengths of opportunity sampling
- quick to gather participants
- very easy as no adverting or selection processes required.
Weaknesses of opportunity samplying
- Biased as based on who is available at the time or who the researchers selects.
- less ethical as participants may feel obliged to take part and continue
Strengths of random sampling
- should be representative of the target population
Weaknesses of random samplying
- Difficult to get all the names of everyone in the target population
- May not be willing to take part (so could still become basised).
Strengths of snowball
- Easy as only requires finding a few participants before they recruit the rest
- cheap as not requiring advertisement
Weaknesses of snowball
- may be more likely to get friends of participants so they could have similar characteristics
- may not get enough participants if they don’t recruit others
Target population
The group of people the researcher is interested in studying
Sampling methond
The different ways in which researchers can obtain a sample of people within the target population to take part in their study
The sample
The actual group of participants used in the research
Primary data
In experiments we are measuring the ability or behavior of each participants within each condition of the experimental; task and collecting the data directly.
Secondary data
Secondary data can be used in other research methods such as correlation, to analyse information which already exists. An example would be crime statistics which the police are already in possession of.
Quantitative data
Quantitative data is about ‘quantities’ of things. They are numbers, raw scores, percentages, means, etc. They are measurements of things, telling us how much of something there is.
Qualitative data
Qualiative data are about ‘qualities’ of things. They are descriptions, words, meanings, pictures, etc. It is data that cannot readily be counted and comes from asking open-ended questions to which the answers are not limited by a set of choices.
Strengths of quantitative data
Easy to analyse statistically
Quick to convert and compare
Weaknesses of quantitative data
Lacks depth or detail about the person or behavior
Lacks insight into personal experiences or explanations of behavior.
Strengths of qualitative
Richer and more detailed on the person or behavior
Provides insight into personal experiences or explanations for behaviour
Weaknesses of qualitative data
Difficult to analyse
more time communing to analyse and compare