Bài giảng Chapter 7 Sampling, Significance Levels, and Hypothesis Testing

Chapter 7Sampling, Significance Levels, and Hypothesis TestingThree scientific traditions critical to experimental researchSampling Significance levelsHypothesis testing1Population and SamplePopulation – all units (people or things) possessing the attributes and characteristics of interestSample -- subset of a populationSampling frame -- subset of units that have a chance to become part of the sampleResearchers study the sample to make generalizations back to the population2Defining the PopulationChoose the dimensions or characteristics meaningful to the hypothesis or research questionMust be at least one common characteristic among all members of a populationMust develop procedure to ensure representative sampling3Addressing GeneralizabilityExtent to which conclusions developed from data collected from sample can be extended to its populationSample is representative to the degree that all units had same chance for being selectedRepresentative sampling eliminates selection bias Characteristics of population should appear to the same degree in sampleRepresentativeness can only be assured through random sampling4Probability SamplingThe probability of any unit being included in the sample is known and equal When probability for selection is equal, selection is randomAlso known as random samplingSampling error will always occur 5Types of Probability SamplingSimple random samplingSimplest and quickestSystematic samplingIf used on a randomly ordered frame, results in truly random sampleStratified random samplingRandom sampling within all subgroupsCluster samplingRandom sampling within known clusters6Nonprobability SamplingDoes not rely on random selectionWeakens sample-to-population representativenessUsed when other techniques will not result in an adequate or appropriate sampleUsed when researchers desire participants with special experiences or abilities7Nonprobability Sampling TechniquesConvenience sampleVolunteer sampleInclusion/exclusion sampleSnowball sampleNetwork samplePurposive sampleQuota sample8Sample SizeNumber of people/units for whom you need to collect dataDetermined prior to selecting sampleLess than the number you ask to participateThe larger the sample relative to the population, the less error or bias9Comparisons of Sample Size to PopulationPopulation SizeSample SizePopulation SizeSample Size100801,0002782001325,00035750021750,00038410Significance LevelsThe researcher sets the significance level, or p, for each statistical testThe degree of error the researcher finds acceptable in a statistical testAn estimate of what would happen if the study were actually repeated many timesGenerally .05 is accepted level11Significance Levels.05 significance level = 5 out of 100 findings that appear to be valid will be due to chanceAlso known as the alpha level or pIf p > .05, the finding is nonsignificantIf p is  .05, the finding is significant or real12Hypothesis Testing Hypothesis states the expected relationship or difference between two or more variablesAlternative hypothesis presented in reportNull is statistically testedAct of decision making based on the significance levelDecision based on comparison between p set before study to p produced by statistical test13Hypothesis TestingBelief in the null hypothesis continues until there is sufficient evidence to the contraryIf p for statistical test exceeds significance level, null is retained (p > .05)If p for statistical test is  .05 then alternative hypothesis is accepted14Error in Hypothesis TestingIn reality, the null hypothesis is trueIn reality, the null hypothesis is falseUse level of significance to reject nullType I error – Null is rejected even though it is trueDecision 1 – Null is rejected when it is falseUse level of significance to retain the nullDecision 2 – Null is retained when it is trueType II error – Null is retained even though it is false15