Statistics Exam 2

What is p value?

the probability of getting a result that is at least as extreme as our experimental result if the null hypothesis was true

what is z score?

a measure of how many standard deviation above or below the mean value a score is

what are the 2 sources of variability?

1) differences due to normal everyday differ btwn individuals aka random

2) differences due to experimental manipulation (of the IV)

This is important because in order to reject the null hypothesis, need the differences by experimental manipulation to be larger than normal, everyday noisy differences within the population

T Statistic

type of parameter test that looks at the difference btwn 2 groups

Formula is the measure of the differ betwn conditions [differ of 2 group means] vs measure of underlying variability in the pop

What is the main difference between parameter statistics and simulating and building null distribution?

With building null distribution the data is from the coin or match applet while for parameter statistics it is using the data that is collected that is often a normal distribution or at least a large sample and using mathematical formulas to build a diagram.

what 2 values needed for parametric statistic?

the mean and standard deviation of the population

How does population variance differs from sample variance?

population variance based on sample data is calculated using n-1 in the denominator, as the sample mean underestimates the variance of the population. the n-1 makes the overall estimate of variance larger.

therefore the population variance is different as the “n-1” represents the degree of freedom aka the # of individual scores are free to vary

What is the File Drawer Effect?

it when research journal only publishes studies that show statistical significance so biased towards positive findings and non significant findings don’t get published

means any published results has higher prob of being type 1 error than just normal 5%

P Hacking

Making small or large changes to the analysis of the data to get significant results

What is effect size and what does it tell us?

how strong the relationship is or in other words how much of an effect (variability for correlations) the change of IV has on the value of DV?

What are confidence intervals and tell us?

how sure we are that the sample results (mean) represents the actual/true population values

{} =smaller the range, higher the confidence

if there is 0 within the range NOT statistically significant

What is a post hoc test?

help us identify after we consider result to be significant, which specific group are different

with anova use bonferroni

what is a between and within groups variance and why do they matter?

HARKing

Hypothesizing after results are known

researcher makes it seem like they had the correct hypothesis all along when they only notice relationship btwn variables in data and create hypothesis after getting the results

What is pre-registration and how can it solve some problem with the scientific process?

this process includes that the researcher hypothesis are, what data they will collect and what analysis they are gonna run. prevents researchers from being biased abt their own results

What is between groups variance (signal)

measuring how much the group means differ from each other

represent effect of the IV variable

what is within group variance? [the noise]

measure the spread of each scores within each individual group : accounts for random error or individual differences

why does between and within group variance matter?

they matter bz the F statistic [anova] is the ratio of the btwn group / within group