Prep for Psyc 201 Exam 2 Flashcards
Highly likely to chance
High p-value
Less likely to chance
Low p-value
Probability of the data given the null hypothesis- NOT the probability that a particular state of the world is true and NOT the probability that the null hypothesis is false
p-value
A statistical method that uses sample data to evaluate a hypothesis about a population. The general goal is to rule out chance (sampling error) as a plausible explanation for the results from a research study.
Hypothesis Test (Two tailed test)
The hypothesis that there is no significant difference between specified populations. Predicts there is NO relationship
Null hypothesis
A statement that directly contradicts the null hypothesis. Predicts there is a relationship.
Alternate hypothesis
the probability value that is used to define the vary unlikely sample outcomes if the null hypothesis is true
significance value (alpha level)
Extreme sample values defined by alpha level, used to reject/retain H0
Critical region
The statistical hypotheses specify either an increase or decrease in the population score
Directional hypothesis test (one-tailed test)
When the probability of certain results are beyond a critical region if the null hypothesis is true. It is determined by the p-value.
Statistical significance
Concerned with whether the results are used in the real world. It is determined by effect size.
Practical significance
Gives us an idea of how large, important, or meaningful a statistically significant effect is.
Effect size
Sample mean larger than the hypothesized mean
Positive value of Cohen’s d
Sample mean smaller than the hypothesized mean
Negative value of Cohen’s d
rejecting the H0 when it is actually true. Affected by the alpha level. Alpha level is the probability of rejecting the H0 when it is actually true. Smaller alpha level, smaller risk of false positive
Type I Error (false positive)
Failing to reject H0 when it is actually false and H1 is true. Affected by the power of the study
Type II Error (false negative)
Defined by the probability that the test will reject the null hypothesis when the treatment does have an effect.
The power of hypothesis testing
Larger sample size, increased power to detect an effect
Sample size affecting power
Larger alpha level, less restriction in detecting an effect
alpha level affecting power
use of statistics in which hypotheses are based on or informed by one’s results.
Hypothesizing After the Results are Known (HARK)
An exploitation of data analysis in order to discover patterns which would be presented as statistically significant, when there is no underlying effect.
p-hacking
Probabilities are calculated based on a sampling distribution. Doesn’t require knowing the population standard deviation; instead, we use an estimate of s. The estimate becomes more accurate with larger sample sizes. Different curves = different critical values. Higher degrees of freedom = more similar to the normal curve
t Distribution
Comparing one sample mean to a predicted population value.
One-sample t-test
Calculated with the estimated standard error
t-value
The margin of error around our point estimate. Can only be calculated for two-tailed tests.
Confidence interval
Less error, more precise
Narrow CI’s
More error, less precise
Large CI’s
X-bar +/- t*(s/sq root of n)
Confidence Interval Formula (Two-tailed test)
An experimental design in which subjects are exposed to all levels of the independent variable and their score son the dependent variable are compared. A single sample of individuals is measured more than once on the same dependent variable.
Repeated measures
Comparing the mean difference between two measurements from one sample.
Paired samples t test
A response to some manipulation or natural occurrence
Changes and Differences
Individuals that are compared within groups, not between groups. Ex: Spouses, siblings, twins
Matched or paired individuals
compares the sample means of two independent groups in order to determine whether the associated population means are significantly different among a single variable
Independent measures
Comparing two sample means between independent groups from one point in time
Independent samples t test
Difference between group means +/- margin of error
Confidence Interval Formula (Independent Samples t test)
The combined estimate of variance using the information from each sample
Pooled variance estimate
t = (Sample mean G1 - Sample mean G2)/ estimated standard error
Independent samples t test formula
t = (mean difference - hypothesized mean difference)/ estimated standard error of mean differences
Paired samples t test formula
t = (sample mean - population mean)/estimated standard error
One sample t test formula
Standard pooled variance = (SS1+SS2)/(df1+df2)
Pooled variance formula
SE = sq rt of (pooled variance/n1 + pooled variance/n2)
Estimated Standard Error for Independent-samples t test
