Ch. 10 - Independent Samples t Test
the larger the sample size, the more
accurately it represents the population
an independent t test can also be used to
compare two distinct pop. of ppl who are already different in some way *in other words, it can compare groups with a preexisting difference
when the obtained t value is not further from zero, then the critical value the null hypothesis should
not be rejected (?)
what is the objective of the independent samples t-test?
determine if the difference btwn the two sample means is likely or unlikely to be due to sampling error
degree of freedom formula
df= (n1-1) + (n2-1)
when computing the standard error of the mean difference, the equation calls for using
the sample sizes from each group (?)
which of the following values can never be negative ? (p329)
the standard error of the mean difference
independent samples t-test
uses two samples from the population to represent two different conditions
if the null hypothesis is true, the z for a sample mean, single-sample t test, related samples t test, and the independent t test all expect an obtained value close to
zero
in most research situations, the populations mean difference that is expected if the null is true is
zero
conceptual formula for the independent samples t
(samples' mean difference - populations' mean difference expected if null hypothesis is true) / (mean difference expected due to sampling error)
which alpha value has a lower risk of making a Type II error ?
.01
which alpha value has a lower risk of making a Type I error ?
.05
when computing the numerator of the independent samples t test, the population mean difference (i.e., μ1-μ2) will always be (p.323)
0
the estimated standard error of the mean difference is (p324)
1. an estimate of how different the two samples means are expected to be due to sampling error 2. (?)
an independent t test can be used to compare differences btwn ppl that (p.319)
1. are created by the researchers by providing different IV levels 2. already exist in different populations of ppl
for all of these tests, if the null hypothesis is true, the obtained t or z should be __1__, however if the null hypothesis is false, the obtained t or z should be __2__ ____ _____
1. zero 2. far from zero
researchers need to be very cautious when interpreting null results because
a null result might occur because of a problem with the study's experimental procedure
Levene's test will help you determine
if the two conditions have similar variances or variances that are very different (i.e., if the homogeneity of variance assumption is met or violated)
when determining if an effect size is small, medium, or large, you should
ignore the sign of the computed effect size and use its absolute value
which significance test should you use to determine if the difference btwn two unrelated samples is likely to be due to sampling error ? (p.316)
independent t test
when you are doing a one-tailed t test, the critical region is always on the side of the t distribution that is predicted by the (p.328)
null hypothesis (?)
the numerator of the independent samples t test is the difference btwn
sample mean and a population mean
the independent t test is a ratio of the difference btwn two sample means over an estimate of
sampling error (SEM)
if the experimental group had a higher mean and the obtained t value is in the critical region, you could conclude (p.317)
that the verbal descriptions increased memory scores
whenever you failed to reject the null hypothesis and yet the effect size is medium or large, you should conclude
that your sample size was too small and you should return the study with a larger sample size
you use an independent samples t statistic when (p.321)
the IV defines two independent samples and the DV is measured on an interval/ratio scale (?)
the pooled variance is the average of the
two sample variances weighted by the sample size
whenever the null is not rejected and yet there there was a medium or large effect size, the sample size used in the study
was too small for the statistical test or the effect size to be trusted
when should you use related samples t test ?
when comparing the sample means from either the same individuals measured at two different times or pairs of matched people measured under different conditions
when should you use independent samples t ?
when you need compare to compare two sample means that are unrelated
when should you use single sample t test ?
whenever you want to compare a sample mean to a population mean (or a value of theoretical interest) but you don't the population's standard deviation
Based on the results of this Levene's test
you will choose btwn the t test that assumes equal variance or the one that does not