Week 9 ---- Testing Hypothesis & Degree of Relationship

hypothesis testing purpose

collect the most accurate data impossible

hypothesis

a testable statement that reflects the general research question

used in experimental research

hypothesis testing

• statistics and probabilities used yo show how confidently a hypothesis is supported

• assume that the null hypothesis is accurate, our goal is to reject the null hypothesis

null hypothesis

• there is nothing to find

• no significance in the data, no statistically relevant association between variables

• the assumption for all research

alternative hypothesis

• there is something there

• two kinds

• directional and non-directional

directional alternative hypothesis

• there is a statistically significant association between the variables

• a direction (positive or negative) is specified

non-directional alternative hypothesis

• there is a statistically significant association between the variables

• direct inverse of the null hypothesis

• no direction is specified

statistically significant

• something is likely causing the phenomenon

• the data suggests that the correlation is not random or by chance

• does not prove causation

• the p-value determines whether or not the data is this

probability value / p-value

• ranged 0 to 1

• p < 0.05 indicates statistically significant data

• closer to 0, the better (0 means there is NO chance that the data is by chance/coincidence)

test statistic

• a number that measures the difference between the observed data and the expected data (the expected data assumes the null hypothesis is true, as all research does)

critical value

• threshold of the decision-making cutoff to determine the hypothesis

• for the p-value, the critical value is 0.05

bivariate inferential statistics

• 2 variables tested at a time

• used to predict

• pearson’s correlation test

• chi-square test

pearsons correlation test

• determines line of best fit

• evalutates the strength and direction of a linear relationship

pearsons correlation coefficient

• measure that calculates the strength and direction of a linear relationship

• range: -1 to 1

chi-square test

• used when both variables are categorical

• contingency tables