Statistical Analysis Flashcards
Describe the different types of statistics
Descriptive–summary of raw values, describes a samples characteristics
Inferential–used to infer something about a population based on a sample’s characteristics (answer research question!)
What are measures of central tendency?
Central tendency=one value that best represents entire group of scores. Mean, median, mode
What are measures of variability?
variability=how scores differ/how much each score differs from the mean. Range, standard deviation, variance.
What can a range be used for?
Gives an idea of how far apart scores are from one another
What is a standard deviation?
average amount of variability in a set of scores. Average distance from the mean.
What is variance?
the square of the standard deviation
confidence intervals
a range of values around a mean that are believed to contain with a certain probability (ex 95%) the true value of that statistic
correlation coefficient
measured the relationship between variables. How the value of one variable changes when the values of another variable changes. Range from -1.0 to +1.0 .
Absolute value =strength. +-=direction
What is the strength of correlation for different correlation coefficients?
1=perfect
.7-.9=strong
.4-.6=moderate
.1-.3=Weak
What is direct correlation?
variables change in the same direction (positive relationship)
What is indirect correlation?
variables change in the opposite direction (negative relationship)
What is a type I error?
False positive. Reject null when null is true.
What is a type 2 error?
False negative. Failing to reject to null when null is false.
What is alpha?
the probability of committing a type 1 error. What value is chosen before study is conducted to determine if statistical significance is present .
Degree of risk you are willing to take to have a false positive.
Which type of error is focused more on minimizing?
Type 1 error. Balancing act between each type of error, type one error is more detrimental to make.
What is a p value?
level of significance found after data was collected (probability value). {Probability of getting a false positive.
p
What is statistical significance?
the likelihood a relationship b/w variables is caused by something more than random chance
What are the steps in hypothesis testing (computing t-test statistics)?
- Statement of null and research hypothesis
- Set the level of risk (type 1 error) associated w/ null hypothesis (alpha)
- Selection of appropriate test statistic
- Computation of the test statistic value
- Determination of the value needed for rejection of the null hypothesis (critical value, p value)
- Compare obtained value and critical value
- Decision.
What is a t-test?
compares the differences between 2 means
What are the different types of t-tests?
Dependent/paired (each case in one sample has a corresponding member in the other sample ex: pretest, posttest)
Independent t-test (two independent samples)
What is ANOVA?
analysis of variance. Compares the amount of variability between groups with the amount of variability within groups. F test F(number of groups-1, participants).
1=differences between groups not significant. As F value increases, more likely not due to chance
What is regression?
Statistical analysis, estimates relationship among variables
What is R2?
proportion of variance in the dependent variable that can be explained by the independent variable. Positively biased, use adjusted R2.
What are the types of regressions?
Linear regression: relationship b/w a DV and one IV
Multiple regression:
relationship b/w a dependent variable and two or more IV
effect size
measure of a findings importance. size or magnitude.
what is practical significance?
usefulness or practical application for the real world
treatment efficacy
impact of treatment under well controlled and somewhat artificial conditions
treatment effectiveness
impact of a treatment as it is actually administered in the “real world” of clinical practice. Ex: generalization
WHat is social validity?
effects large enough to be noticed by naive observers
