This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Confounding variable : extraneous variable that correlates with DV and IV Constructs : internal attributes/char that can’t be directly observed Operational : measure external behavior so measurements can be used to infer the status of underlying construct bar chart : discrete data w/ nominal or ordinal scale Line chart : discrete data w/ interval ratio scale Scatter plots: continuous or discrete data w/ both variables of interval or ratio scale Frequency distributions (histograms): continuous data Shape of distributions: symmetry (a/symmetric) and skew (postail right, neg tail left) Measures of central tendency Mean : interval or ratio scale (NOT nominal or ordinal), pop (u=∑X/N), sample (M, X) Median : ordinal/interval/nominal; when data is skewed or there’s extreme scores (outliers) Mode : for all scale data; when most common score is most meaningful way to describe data skew for pos: mode, median (best CT) and mean ERROR TYPES: If z>zcrit REJECT H0. If z<zcrit retain H0 or fail to reject H0 no effect, H0 TRUE: (1) retain H0 – correct or (2) reject H0 – type I error , p=alpha effect, H1 TRUE: (1) retain H0 – type II error, p=beta or (2) reject H0 – correct, p=1beta ( power: The probability that the test will correctly reject a false null hypothesis. That is, power is the probability that the test will identify a treatment effect if one really exists. A false alarm is type I error, which only occurs when null hypothesis is true, but we reject the null hypothesis. On the other hand, a miss beta is type II error, which only occurs when the alternative hypothesis is true, but we fail to reject the null hypothesis. Basically, "false alarm" and "alpha error" are other ways to say type I error, while "miss" or "beta error" are other ways to say type II error alpha level : The probability that the test will lead to a Type I error. That is, the alpha level determines the probability of obtaining sample data in the critical region even though the null hypothesis is true.alpha level (or the level of significance) A probability value that is used to define the concept of “very unlikely” in a hypothesis test. o For a nondirectional hypothesis test, if the alpha level of a study is changed from alpha = 0.05 to alpha = 0.01, the probability of a false alarm will shift from 0.05 to 0.01 assuming that the null hypothesis is true. DEGREES OF FREEDOM (if n=16) : For the repeated measures study, df = (nD 1), thus there were 17 difference scores, which means 17 individuals participated. For the matched subjects study, df = (nD  1), thus there were 17 difference scores; difference scores are between different individuals, thus there were 34 participants. For the independent measures study, df = (n11) + (n2 1), or (ntotal – 2), thus there were 18 individuals total. Mean Standard deviation for population, From SS Standard deviation for sample, From SS Sum of squares for population, defn form...
View
Full Document
 Spring '11
 PARRIS
 Statistics, Normal Distribution, Standard Deviation, Statistical hypothesis testing

Click to edit the document details