Scientific method

Question 1

Major features of science:
Empiricism \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Objectivity \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Replicability \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Control \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Define peer review \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Its 3 main purposes:
Allocating funding \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Publishing accurate, non-faulty data \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Assessing uni departments \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_

Answer 1

Empiricism: information is gained via observation/experiment, not just argument/beliefs.
Objectivity: expectations should not affect what is/is not recorded.
Replicability: experiments should have the same outcome when repeated.
Control: ensured with the experimental method - varying the independent variable to observe its effect on the dependent variable, and controlling any and all other variables there may be.
Peer review: assessment of scientific work by experts from the same field as the researcher
Allocating research funding - it is paid for by government/charity bodies who need to decide which research is worth spending on.
Publishing data: journals in which data is published allow scientists to share this information, and if it is peer reviewed the data should not be faulty/incorrect.
Assessing departments: university departments are assessed on the quality of their research and department funding depends on good ratings from peer review.

Question 2

Independent variable ___________________
Dependent variable ________________

Experiments:
Laboratory \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Field \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Natural \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Experimental designs:
Repeated measures \_\_\_\_\_\_\_\_\_\_\_\_
Independent groups \_\_\_\_\_\_\_\_\_\_\_\_\_\_
Matched pairs \_\_\_\_\_\_\_\_\_\_\_\_\_

Answer 2

Independent variable: what the experimenter varies in order to observe its effect on the dependent variable.
Dependent variable: what is measured in terms of how the independent variable has affected it.

Laboratory experiments involve a controlled environment, therefore high replicability and internal validity but low external validity.
Field experiments are conducted in a natural environment, so have reduced experimenter effects but reduced control of extraneous variables.
Natural experiments use pre-existing independent variables, so causal conclusions can’t be drawn, but natural experiments must be used for certain behaviours.
Repeated measures involves testing each participants on all IVs
Independent groups involves having separate groups for each IV
Matched pairs is the design with separate groups but each group contains a participant matched with a participant from the other group(s) e.g. matched in intelligence.

Question 3

Reliability:
In experiments \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
In observations \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
In self-report methods \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Validity:
In experiments \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
In observational studies \_\_\_\_\_\_\_\_\_\_\_\_\_\_
In self-report methods \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_

Answer 3

Reliability:
In experiments reliability is how well the study can be replicated to obtain the same result.
In observations, inter-rater reliability is used: a calculation of the extent to which two separate observers agree in observations.
In self-report, internal reliability measures the consistency of the method with itself e.g. whether IQ test questions all measure IQ in the same way, and external reliability is measuring the method’s consistency across more than one separate occasion.
Validity:
In experiments, internal validity (whether the researcher tested what they intended to) and external validity (whether the results can be generalised to other situations outside of the study) are affected by the type of experiment (laboratory/field) but also in general whether the participants knew they were being studied and whether the task itself was artificial.
In observation, validity depends on the coding system (relevant/accurate categories) and removing any observer bias.
Self-report methods should obviously measure what the researcher intends to measure and concur with similar tests in order to be valid.

Question 4

Statistical tests (Spearman’s Rho, Chi-square, Wilcoxon T and Mann-Whitney U)
For all, significance level is usually p≤____ and the test is non-p______
The type of data will be o______ or i_______ except for Chi-square which is used when there is n_____ data
The type of design will be i__________ if Mann-Whitney is used, otherwise it will be r________________
When there is an R in the name _____________________________

Answer 4

Significance level is usually p≤0.05
The test uses non-parametric data
Ordinal or interval for any, except Chi-square where it is nominal data.
The design will be independent measures if the test is Mann-Whitney, otherwise it will be repeated measures.
When the test has an R in the name the observed value should be gReateR than the critical value, otherwise it should be lower.

Question 5

Self-report methods:
Questionnaires and interviews _______________________

Observational studies: ______________

Correlational analysis: \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Data types: 
Nominal \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Ordinal \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Interval \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_
Ratio \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_

Answer 5

Questionnaires and interviews use questions to find out what people think/feel with the main problem being honesty of responses (social desirablility bias)

Observational studies simply involve watching behaviour and recording it within behavioural categories. This gives a rich picture of what people actually do but may be affected by observer bias - the effect of an observer’s expectations.

Correlational analysis identifies where there is a relationship between co-variables.
Nominal data is organised by category/name
Ordinal data is organised into a specific order, but without specified intervals.
Interval data: there is a specified interval/unit of measurement separating the data.
Ratio is interval data but with a fixed zero point.