# Stats Flashcards

Terms Definitions
 Direct relationship In this kind of relationship, those who score high on one variable tend to score higher on the other, & vice versa Positive relationship See direct relationship Indirect (inverse) relationship Those who score high on one variable tend to score high on another. Negative relationship See inverse relationship Causal relationship Est correlation does not necessarily establish causal relationship. Pearson product moment correlation coefficient The full, formal name of pearson's r. Ranges from -1.00 to 1.00. Both -1.00 and 1.00 indicate a perfect relationship. Interpretation of pearson's r Varies depending on type of study. Pearson's r Not a proportion. Thus multiplying it by 100 does not yield a percentage. Test reliability One way to examine test reliability is to administer the same test twice to a group of participants without trying to change the participants between administrations of the test - then correlate. Coefficient of determination r squared. When r2 is converted to a percentage it indicates how much variance on one variable is accounted for by the variance on the other. Variance. Spread Variance explained r2 x 100. Aka variance accounted for. Scattergram Illustrates correlation between two variables. Outliers Exception to the overall trend. Multiple correlation coefficient The symbol for a multiple correlation coefficient is upper case italicize R. A multiple correlation coefficient has the same basic characteristics as r. Determines the degree of relationship between a culmination of two predictors. Linear Regression A Regression analysis that assumes that the predictor variable is related to the criterion variable through a linear function Truncated range A range of cases that lacks one or both ends of the distribution of values Restricted range Limitation by researcher– Via sampling, measuring Procedures, or other aspects of experimental design – of the full range of total possible scores that may be obtained to only a narrow, limited portion of that total Curvilinear correlation A functional relationship between variables that is not a straight line form when depicted graphically Random sampling -yields an unbiased sample; names can be drawn to obtain a random sample; a table of random numbers may also be used to draw a random sample; to use a table of random numbers give each person in the population number name Stratified random sample Draw participants at random separately From each stratum Multistage random samples May be used in large-scale studies Random cluster sampling Existing groups of participants are drawn. All members of population must belong to a cluster Sampling Errors A random sample may differ from a population in important aspects by the luck of the draw Sample size The larger the sample, the smaller the sampling errors and the greater the precision; increasing sample size produces diminishing returns; the smaller the anticipated difference in the population the larger the sample should be; for populations with very limited variability even small samples can yield concise results. The more variable the population the larger the sample size should be. When studying a rare phenomenon large samples are usually required. Large samples do not correct Four bias Precision The larger the sample the smaller the sampling errors and the greater the precision Central limit theorem According to this theorem the sampling distribution of Means is normal Standard error of the mean The standard deviation of the sampling distribution is known as the standard error of the mean Sampling distribution of mains A large number of means Margin of error The standard error of the main is a margin of error Confidence interval for a mean Confidence interval assist and interpreting means that are subject to random errors. they cannot take bias into account. Using large samples keeps the standard error of the mean and the confidence interval small Point estimate A single estimated numerical value of the population parameter Interval estimate An estimate range of likely values for a given population parameter Observed difference The difference a researcher Obtains. It may not represent the true difference because of the influence of random sampling error. Significance tests Test of the null hypothesis P value Significance level. The probability of rejecting the null hypothesis when it is in fact true. – Such as making a type one error Alpha level The probability level for rejecting the null hypothesis Decision error Type one error: reject the null hypothesis when it is true – type two error: felt to reject the no hypothesis when it does false About the population Testing the null hypothesis about the population One tailed significance test For directional hypothesis Two-tailed significance test nondirectional hypothesis T test Tests the difference between two sample means for significance Statistically significant Hey synonym with rejecting the null hypothesis Not significant (n.s.) Failure to reject the null r Symbol for Pearson product moment correlation coefficient
/ 45
Term:
Definition:
Definition: