Research methods

Question 1

Confounding vs extraneous variables

Answer 1

EV are apart from IV than could affect DV
CV have already affected the DV

Question 2

Operationalisation

Answer 2

Defining a variable and stating how it will be measured

Question 3

Lab experiment

Answer 3

Very controlled environment and artificial setting where participants are randomly allocated to a condition
+ High control to eliminate EV so CV, easily manipulate IV to establish cause and effect, easily reproducible for reliability
- demand characteristics, social desirability bias, mundane realism, ecological validity

Question 4

Field experiments

Answer 4

real world or natural setting though IV still controlled
+ mundane realism, ecological validity, IV can be controlled to establish cause and effect between IV DV, less demand characteristics esp if unaware they are taking part
-less control over EV - maybe invalid, less control of sample so less representative, difficult to reproduce

Question 5

Natural experiment

Answer 5

uses naturally occurring IV not directly manipulated, finding participants who already meet the conditions of an experiment
+ mundane realism, ecological validity, useful when impossible or unethical to manipulate IV in a lab/ field exp
- low control over EV, very difficult to reproduce, difficult to determine cause and effect due to low control

Question 6

Quasi experiment

Answer 6

Naturally occurring IV which is a difference between people that already exists - usually in a lab setting
+ High control, less EV, easy to reproduce with control
- Lacks ecological validity as artificial setting, demand characteristics, validity

Question 7

Behavioural categories

Answer 7

Specific observable behaviour groups recorded in an observation - to be operationalised
Easier to focus, tally, objective, quantifiable, analysis, reliability

Question 8

Observer bias and fixes

Answer 8

Knowing the aims of the study will influence observations, or observations could be very subjective
Inter-observer reliability - 2 people watch and record then correlate via Kappa test where +0.8 indicates reliability
Intra-observer reliability - recorded so can be rewatched several times (and by others)

Question 9

Sampling in an observation

Answer 9

Event sampling - recording every time something happens - useful to gain a full picture and not miss anything out, but sometimes hard to count esp if very frequent
Time sampling - recording in given time intervals - easier esp in hard to count situations, but risks missing out major events between intervals

Question 10

Types of observations

Answer 10

Non participant - no involvement with participants
Participant - direct involvement with actions/ interactions of participants
Covert - undercover and may give fake identity, group unaware
Overt - self introduce makes clear that they are being observed by others

Question 11

Naturalistic vs controlled observation

Answer 11

Naturalistic observes in own environment with no direct IV manipulation - usually covert
Controlled observes in artificial environment with manipulated EV - usually overt
Manipulating EV makes it easy to establish cause and effect but may be difficult/ unethical
Naturalistic usually more ecologically valid and mundane realism

Question 12

Generic observation eval

Answer 12

Whether people know - ethics, observer effects, demand characteristics, social desirability bias, mundane realism, ecological validity
Researcher participation - focus on observing, relatability to actual activity
Observer bias (can be overcome)

Question 13

Interviews

Answer 13

Structured - same questions in same order, closed questions for quantitative data
Unstructured - informal in depth unplanned, open questions for qualitative data
Semi structured interview - mix of both

Question 14

Interviews eval

Answer 14

+Unclear questions can be clarified easily,
- interviewer effects can occur - even with uniform, body language
+ Can be recorded to replay/ analyse
- Need training/ skill esp for open qs - taking notes makes it harder to focus but not taking notes makes the participant feel unheard
- Social desirability bias
- Structured are quicker but lack in depth data, opposite for non structured
- Unstructured questions may lead to off topic/ useless discussion
Semi structured combines + and - of both

Question 15

Questionnaires + design

Answer 15

Written set of questions/ instructions - can be self administered, by post, face 2 face etc
Consider type of data - qualitative to open qs and vice versa
Avoid ambiguous questions
Avoid questions with two parts
Avoid leading questions
Avoid complex language so easily understood

Question 16

Questionnaires eval

Answer 16

+ Easier, quicker, cheaper to get large sets of data
+ Easy to replicate to establish reliability
- Ambiguous questions cannot be clarified
- low response rate esp if needs to be reposted or people forget about it/ miss some questions
- participant variables as mainly less busy or older people answer them

Question 17

Correlations

Answer 17

analyse strength and direction between co variables from -1 to +1
plot covariables on a scattergraph

Question 18

Correlation eval

Answer 18

+Establish strength between 2 variables and measure precisely
+Predictions can be made based on correlations
+Can investigate things that cannot be manipulated experimentally for ethical or practical reasons
- Cause and effect unclear
- Third variables - correlation and causation
- Doesnt detect curvilinear relationships - positive until a certain point then negative (e.g enzymes) or vice versa

Question 19

Aims

Answer 19

Precise statement about the purpose of the study and what it intends to find out
e.g an investigation to study the effects of eating chocolate on mood
should lead to operationalised variables

Question 20

Hypotheses

Answer 20

Null - no significant effect
Alternative/ experimental hypothesis - will have an effect, either one tailed or 2 tailed
Results are significant if statistically unlikely due to chance