basic laboratory experiments - reading chapter 9 Flashcards
Repeated measures design
same ppt + all conditions – correlated t – test
Ppts:
• Condition 1
• Condition 2
Independent groups design
different ppt + different condition – unrelated t-test
Ppts:
• Group 1
- condition A 1st
- condition B 2nd
- Group 2
- condition B 1st
- condition A 2nd
Characteristics of true or randomised experiment – aspects of experience
- Experimental manipulation
- Standardised procedures
- Random assignment
• Experimental manipulation
- manipulation of IV to test effects on DV
- condition with lower level of IV = control group e.g. lower alchohol
- condition with higher IV = experimental condition
- experimental control condition should be almost identical aside from IV
• Standardised procedures
- all factors should remain constant aside from IV (standardisation)
- lower variation in extraneous variables (e.g. time of day)
- lower/ no systematic bias for one of the conditions
• Random assignment
- 2 main procedures
- ppts assigned to experimental or control group using proper randomising procedure
- only used when ppts take part in 1 condition
- ppts randomly assigned to complete conditions in different orders
- only used when ppts take part in more than 1 condition
random
(to draw conclusions about causality) – each possible outcome has equal chance of being picked
- Possible procedures: toss a coin, throw a die, computer generation
matching
ensuring ppts in experiment . Control groups are similar on variabes which might affect results (e.g. energy, IQ)
Pre-test post-test sensitisation effects
- Checks if random assignment has equated groups on DV prior to experimental manipulation. So it is clear that different results due to manipulation of IV or pre-existing differences.
- Allows to test if there was a difference in performance between pre and post-test/
Within subject design/ independent groups
- Some ppt different/ all conditions
(-) fatigue/ boredom/ predict a practice
-combat by counterbalancing
Statistical significance
a determination that a relationship between two or more variables is caused by something other than chance. Statistical significance is used to provide evidence concerning the plausibility of the null hypothesis, which hypothesizes that there is nothing more than random chance at work in the data.