DESCRIPTIVE STATISTICS Descriptive Statistics: we are interested in the analysis of sample data only. Technically the term statistic refers to sample data only. No attempt is made to generalize the findings of a research study to a larger population group. For example, a teacher compares three spelling methods for the purpose of determining which method is most effective with her second graders. She plans to use the results ONLY with her second graders – Descriptive use – no attempt is made to extend the results to a larger population of all second graders. (ActionResearch, Evaluation & Assessment Studies). Inferential Statistics: Using sample data to infer population parameters (refers to numbers that describe the population like the mean, median, or mode, range, standard deviation). Take a sample of the larger population group conduct an experiment for the purpose of determining which method of spelling instruction works best (sample MUST be representative of the larger population group so that the researcher can generalize his or her findings). Population: Are the people to whom you wish to generalize the results of any research study. Sample: Refers to a group of individuals that are necessarily smaller than the population from which they were drawn. Representative Samplingis an important element of research. Methods of choosing a representative sample: Simple random sample: The definition is that all S’s in the population have an equal and independent chance of selection.
Stratified Random Sample: Divide the population in to subgroups reflecting population densities and demographics and then sample proportionately. Cluster Sample: For example the use of an intact class as the unit of analysisor school level data (county level data). Nonprobability sample(convenience sample): Use of volunteers – no attempt is made to select a representative sample from the population. Variable: Construct that can be rendered both observable and measurable – such as anxiety, motivation, scholastic ability, or achievement. Can be physical like height or weight. Method of Reading Instruction. Value: Implies group membership, ie, phonics, sight word, or whole language. Research Question: Is there a significant difference in method of reading instruction (phonics, sight word, whole language) on reading proficiency (Reading Running Record Scores)? Independent Variable(Grouping, Factor, Predictor): Is Method of Instruction(Phonics, Sight Word, Whole Language). Dependent Variable(Criterion, Outcome): Reading Skill(Reading Running Record Test). Levels of Measurement: Nominal(Categorical, Grouping, Factor) – Naming variable like eye color or hair color.
Ordinal– Implies a hierarchy where one level is greater than another. For example instructors, assistant, associate, and full professors.