Download Repeated Measures Design: Within-Subjects Experiments and more Quizzes Psychology in PDF only on Docsity!
Repeated Measure Design
(within subjects) Progressive Error Carryover effect Counter balancing: how to control for progressive error and carryover effect Within-Subjects Factorial Designs= within subjects design with more than one factor Mixed Design= design in which one factor is between subjects and one factor is within subjects TERM 2
Between Subjects (independent group) design
DEFINITION 2 each subject served in only one condition of an experiment TERM 3
Within- subjects (repeated measures) design
DEFINITION 3 each subject served in multiple (perhaps all) conditions of an experiment TERM 4
repeated measures design
example
DEFINITION 4 Hypothesis: does depth of processing affect how well information is remembered?Depth of processing- the way in which a person thinks about info shallow= simply read a word and get it deep- think about the word means, think of a word that is similar or rhymes TERM 5
advantages of repeated measure
design
DEFINITION 5 requires fewer subjectsbetter control of extraneous subject variables- individual differencesStatistically more powerful- better chance of detecting the effect
disadvantages of repeated measure
design
Practicallimitations- may require subject to be tested for too longProgressive errorCarryover effect TERM 7
progressive error
DEFINITION 7 as the experiment progresses results are distorted in some way. Includes the following order effects fatigue effects practice effects TERM 8
order effect
DEFINITION 8 change in subjets reponse that occurs when a condition falls in different positions in a series of treatmentsex.pepsi, diet pepsi, coke, or diet cokediet coke might not be as satisfying becuase they arent thirsty by then TERM 9
fatigue effect
DEFINITION 9 subjects could get tired, bored, or irritated TERM 10
practice effect
DEFINITION 10 An improvement in performance as the repeated practice with the task
Randomized partial counter
balancing
randomly select out as many sequences as there are subjects for the experiment TERM 17
Latin Square counterbalancing
DEFINITION 17 a matrix of sequences is constructed that stisfies the condition that each treatment appears only once in any order position in the sequence TERM 18
balanced latin square design
DEFINITION 18 control for both order and carryover effect TERM 19
within subject factorial design
DEFINITION 19 completely within subhect design: Each subject is exposed to each combination of factorsfactor one has 3 itemsfactor 2 has 4 itemseach subjject is exposed to 12 treatments TERM 20
mixed design
DEFINITION 20 one of more factors are repeated, one or more factors is independentlike a phylogeny chart
Statistics
three basic steps to analyzing data organize it summarize it-descriptive statistics Apply a statistical test- inferential statistics TERM 22
Raw data
DEFINITION 22 All of the data recorded while the experiment is being runAny given experiment generates a lot of data and no one wants to read thatinstead the experimenterreportssome combination ofdescriptiveand inferential statistic TERM 23
descriptive vs. inferential
DEFINITION 23 Descriptive- Describes data obtainedInferential- used to draw inferences about the larger population for which the sample is thought to represent. TERM 24
frequency distribution
DEFINITION 24 graph that indicated the number of individuals that recieve different scores on a variable.purpose: to get an idea of the shae of the distributionmethods of graphing: pie chart bar graph frequency histogram frequency polygons TERM 25
continuous vs discrete variables
DEFINITION 25 Continious- a variable that can take on any value within a range100-200200-300Discrete-a variable that can only ake the specific levels o categoy1st grade, 2nd grade...
normal distribution
Majority of the scores are in the middleSymmetrical TERM 32
Positive skew
DEFINITION 32 AsymaetricalThe tail at the higher end of the scale is londer than the lower endlllllllllllllllllllllllllllllllllllllll TERM 33
negative skew
DEFINITION 33 l ll llllll lllllllllllllllllllllllllllllllllllll TERM 34
bimodal distribution
DEFINITION 34 two distinct peaksroughly assymetric TERM 35
mean median
DEFINITION 35 the mean is sensetive to extreme scores so it is better to use the median
mean median mode for positive skew
mode< median < mean TERM 37
mean median mode for negative skew
DEFINITION 37 mean < median < mode TERM 38
variability
DEFINITION 38 Range - the difference between the highest and lowest Variance - the average squared deviation of scored form their mean, a measure of how spread out the data is Standard deviation - the average deviation of the scores about the mean TERM 39
inferential statistics
DEFINITION 39 hypothesis testingtype 1 and 2 errorsfactors that increase power TERM 40
statistical inference
DEFINITION 40 making a statement about the population based on the date obtaned from a sample
hypothesis
testing
what is the probabilyu that the results is significantly different from what would be expected given the usual variabilty amoung the population?ex. backround music has an effect on the productivity rate of workers in a mail sorting room TERM 47
Hypothesis testing step 1: the null
hypothesis
DEFINITION 47 scored from the two treatment conditions are so similar that any differenced were siply due to chance TERM 48
Hypothesis testing step 1: alternative
research hypothesis
DEFINITION 48 data from the two samples are so different that is it highly unlikeley that these differences could be simply due to chance TERM 49
Hypothesis testing step 2: choosing a
significance level
DEFINITION 49 choosing a significance levelsignificance level a=.05we will reject the null hypothesis if the data we have are so different that the chance of us getting these data is only 5% TERM 50
Hypothesis testing step 3: compute the
statistical test
DEFINITION 50 t-test- can be used to test whether two groups are significantly different from each other
degrees of freedom (df)
the number of values in the calculation of a statistic that are free to vary=(# of subjects per condition -1 )(# of conditions)if we have 30 subjects in each group...Df= (30- 1)2= TERM 52
Hypothesis testing step 3: compute the
statistical test
DEFINITION 52 t=mean(group 1)-mean(group 2)/standard deviation / SQRT(number of subjects)4-3/ (2/SQRT(60)=3.8the larger the difference in group means, the larger the t statistic will be TERM 53
step 5: make a
decision
DEFINITION 53 if obtained > critical then rejects TERM 54
type 1 error
DEFINITION 54 a man fails a breathalizer even though he is sobermore severe TERM 55
type 2 error
DEFINITION 55 a dr fails to find a tumor on a x-ray checking for tumors
non-parametric test
a statistical test that does not rely on data belonging to any particular distribution TERM 62
normality
DEFINITION 62 the assumption that the population is normally distributed TERM 63
equal variances
DEFINITION 63 the assumption that the samples have equal variances TERM 64
independence
DEFINITION 64 two groupd are independent if the occurence of one event has no influence on the likeihood of occurance on the second eventtechnically must sample with replacement to have independence however rarely is practices TERM 65
assumptions
DEFINITION 65 manny statistical tests are robust to mild violations of assumptionsKeselman found that researched rarely check that their data satisy the assumptions of the analyses methods used
t-test
any statistical hypothesis test in which the test statisitc follows a student - distribution if the mull hypothesis is suppoerted TERM 67
most common used of t-
tests
DEFINITION 67 two independent samples t-test paired samples t-test one-sample t-test slope of a regression line TERM 68
independent samples t-
test
DEFINITION 68 used to compare means from independent groupsassumptions normality equal variances indepenence TERM 69
paired samples t-
test
DEFINITION 69 two matched group designone way, repeated measures with 2 levels of IVex. hypothesis- do competitive events increase testosterone leels in amles?experimental condition- competitive gamecontrol condition- cooperative game
