100%(1)1 out of 1 people found this document helpful
This preview shows page 5 - 8 out of 9 pages.
Main differencebetween METRIC and CATEGORICAL data-Metric: data that you can add, subtract, and computeaverages for-Categorical: can NOT add, subtract, and compute averagesforScales-primary scales:1. Categorical1) Nominal公稱: no ranking-ordering between values2) Ordinal序數: ranking ordering between values2. Metric1) Interval間 隔: most measures of the strength of feelings,emotions, attitude and beliefs, or extent of behavior are basedon an interval scale.2) Ratio比 例: use ratio scale to provide exact quantitymeasures of tangible thingsWk5Sampling-the process of choosing a sample for your research project-objective: the sample should be representative of the targetpopulation of your study.Sampling processfor your survey1.Define your target population1.1Choose a sampling approach1.2Specify the sampling frame
1.3Specify the survey method1.4What is the sampling technique?1.5Determine the sample sizeSampling TechniqueUse probability sampling techniques-Effective in reducing sampling error-Less subjective-Results conductive to statistical analysis-Findings easily generalizable to the whole population1) Simple Random Sampling簡單隨機抽樣: each element inthe population has a known and equal probability of selection.2) Systematic Sampling系統抽樣: chosen by picking everyi-th element in succession from the sampling frame.3) Stratified Sampling分 層 抽 樣: first divide targetpopulation into groups, picking elements from eachgroup usinga probability-based procedure.4) Cluster Sampling整群抽樣: divide target population intogroups, each group by itself is representative of the wholepopulation, randomly pick one cluster, either use the pickedcluster to be your sample or pick a sample from the cluster usinga probability-based procedure.Coding-means assigning a code, usually a number, to each possibleresponse to each question-ways to code the response comes down to convenience andconventionGet to know your data-target population-sample size-information has been collected-describing individual variables (categorical, metric)*”average” attitude/behaviour and average effectApplies to all comparative RQs, and true for many rationalRQs.Re-coding -need to make changes to the collected data for yourresearch-two types of re-coding that you would encounter in this unit1) Creating new variables that allow you to classify/groupyour responses differently
2) Use a different set of values to code an existing variableHypothesis testing1.Problem definitione.g. do male and female students differ on average intheir grade expectation for MKF2121?2.Clearly state null and alternative hypothesisNull: H0-No difference/relationshipAlternative: H1-difference or association is expected.3.Choose the relevant statistical testTest of association-relational RQTest of difference-comparative RQ4.Calculate p-value5.Is p-value<0.05?No-do not reject nullYes-reject null hypothesisWk8Tests of difference-t-test1.one sample: Compare the averagevalue of onevariable toa content2.paired samples: Compare the averagevalue of twovariables3.two independent samples: H0: µ1=µ2 H1: µ1≠µ2-ANOVAMore than two samples: e.g. Do students in differentfaculties differ in their average marks?