statistical significance and intuitive null hypothesis testing

judging whether result is STATISTICALLY SIGNIFICANT just by looking at it

sampling types

individuals chosen RANDOMLY from population, almost NEVER used in psych research

people chosen because they are "convenient", almost ALWAYS used in psych research

people chosen, usually from convenience sample, and they send suvery to EVERYONE THEY KNOW. similar to convenience sample

different random samples from same population will produce DIFFERENT RESULTS, no "reason" for this, its just random variation

used for INFERENTIAL STATS, focuses on simple binary decision, does NOT tell us how BIG the DIFFERENCE is or how STRONG a correlation is

if i think theres SOME relationship in population, then sample result is

if i think theres NO relationship in population, sample result is

theres NO REAL RELATIONSHIP in population, theres no STATISTICALLY SIGNIFICANT RELATIONSHIP, we either REJECT or FAIL to reject it, NEVER accept it. its the STATUS QUO.

there IS REAL RELATIONSHIP in population. there IS a STATISTICALLY SIGNIFICANT RELATIONSHIP, its what we are TESTING that there IS a relationship.

means that there IS a relationship, our results BACK UP what we though would happen ( H1 is "correct")

means that theres NO relationship, results do NOT back up what we thought would happen, status quo REMAINS. ( H0 is "correct")

STRONGER relationships more likely to reflect REAL relationships in population

results based on LARGER samples are more likely to reflect REAL relationships in population

DETERMINES if relationship is STATISTICALLY SIGNIFICANT or not

sample result is STRONG ENOUGH and based on LARGE enough samples to convince us that there IS a REAL relationship in population, result DID NOT just happen by CHANCE, does NOT mean that the results are big

if p<0.05

if p>0.05

we are 95% confident in our NULL HYPOTHESIS test results

your sample may NOT be representative of the population

note about very large samples

tells us the direction of the relationship, (positive or negative) and the strength (weak/small, medium, strong/large). a significant relationship does not mean it’s large.

Used to make an educated guess about what is true in the population