2Learning Objectives Review Lectures 8 and 9 Understand and develop research hypotheses and know difference between them and the null hypothesisDefine independent and dependent variables for a research hypothesisDefine probability and describe it’s relationship to statistical significance
3Review of Lecture 8Defined and discussed the theory and rules of probabilityCalculated probability and created a probability distribution with example dataDescribed the characteristics of a normal curve and interpreted a normal curve using example data
4Review from Lecture 9Defined research hypothesis, null hypothesis and statistically significanceDiscussed the basic requirements for testing the difference between two meansDefined and described the difference between the alpha value and P value, and Type I and Type II errors
5Research HypothesesHypotheses give a testable and potentially falsifiable prediction about the relationship between two variables.Designed to answer a research question of particular interest.For example, in our child protection study, parent or carer stress was predicted to be significantly associated with the quality of the family environment. This was a central hypothesis.Our research question was: Is parent or carer stress associated with the quality of the family environment?
6Research and Null Hypotheses RESEARCH HYPOTHESISA proposed explanation for a phenomenon that can be testedThere is a relationship between two measured variablesA particular intervention makes a difference/has an effectNULL HYPOTHESISThe opposite position of the hypothesis (usually)There is no relationship between two measured variablesThe particular intervention does not make a difference/has no effect
7Research and Null Hypotheses Examples RESEARCH HYPOTHESISSymbolized as “H1”Parent or carer stress will be significantly associated with the quality of the family environment.NULL HYPOTHESISSymbolized as “H0”Parent or carer stress will not be significantly associated with the quality of the family environment.
8Alternative Hypotheses Alternative or rival hypotheses may offer another explain on why two variables may or may not be associatedAlternative hypotheses are based on the information that you may not have collected or didn’t consider for every possible variableOther variables can:Be the actual causeAlter the relationship between the two variablesIt is important to read prior research literature before doing your research and data collection
9Independent and Dependent Variables Independent variables (IV) those variables of interest which are used to predict dependent variables (DV)Independent variables are also called “Predictors”.Dependent variables are also called “Outcomes”.That is IV explain variation in DV.For example, parent or carer stress (IV) was predicted to be significantly associated with the quality of the family environment (DV).
10ProbabilityResearch and quantitative tests produce results in probabilisticProbability that the association found between an IV and DV occurred due to chanceCan also be said that the association between the IV and DV was statistically significant, and therefore not due to chance
11Statistical Significance In order to determine if something is statistically significant, you must establish a level of significance (represented by the Greek letter α [alpha]).α = the level of probability where the null hypothesis can be rejected with confidence and the research hypothesis accepted with confidenceA common level of significance α = .05
12Statistical Significance In statistical analyses, we find the p-value of the association between two variables (IV and DV).If the p-value is less than our α = .05 level of significance, when we reject our null hypothesis and accept our research hypothesis.If the p-value is greater than our α = .05 level of significance, when we say that we retain or fail to reject our null hypothesis.