Statistical Analysis and Power Calculations: Sample Size Estimation and Research Design Considerations

Page 1 of 4

Statistical Analysis and Power Calculations: Sample Size Estimation and Research Design Considerations - Page 1 preview image

5/27/2015StatisticsDeaconIceLion4113Course Name

Page 2 of 4

Statistical Analysis and Power Calculations: Sample Size Estimation and Research Design Considerations - Page 2 preview image

Page 3 of 4

Statistical Analysis and Power Calculations: Sample Size Estimation and Research Design Considerations - Page 3 preview image

Solution1.Compare and Contrast Internal and External ValidityInternal ValidityExternal ValidityThe internal validity depends over theaccuracy of the results.Therefore, if thesample didnot selectrandomly, then itaffecttheinternalvaliditybecauseinternal validity depends over the datacollection method.Theexternalvalidityincludestheconcepts of generalization to know theimpactsoftheresultsforthelargerpopulation (Herek, 2012). Hence, it alsodepends over the data collection methodas like internal validity.Theinternalvaliditygenerallymakessomedealing with theresearch studywithoutthehelpofotherresearchelements.Externalvaliditylooksoverthatsituation,whenonecouldtaketheoutcomesoftheresearchstudyandsimplify it in a broader perspective. Thegeneralization also works with the helpof external validity (Kimmel, 2007).Itcouldbebetterunderstoodwithanexample. When researchers conduct aresearch study for the testing of drug,the drug organization could not consistinthetestpopulationasthetargetpopulation.Let’s take an example:“To identify theimpactofchangesinecologyontheliving style of teenagers Hispanic” then itwould be very difficult to generalize andsimplify the outcomes of this question forthewholepeopleoftheHispaniccommunity.Inthissituationinternalvalidityisaprimaryconcernbecausewithout it, the result of the study couldnot be effective (Herek, 2012).Let’s take an example:“To identify theimpact of changesin ecology ontheliving style of Hispanic”. Then, for thisresearch question, the external validitycould be a primary concern because thisquestion focuses over the whole groupofHispanicinsteadofsomeoneoraparticular group.For making strong claims in the context of applicability of findings towards a targetpopulation the researchers use the random sampling method, systematic samples,and cluster sampling for gathering the information and data from the selectedpopulation (Kimmel, 2007).This strategy plays an effective role in the generalizationof study outcomes because the sampling process based onprobability alwaysincludes the high response rate that helps the researchers to apply the result ofstudy in the broader manner.

Page 4 of 4

Statistical Analysis and Power Calculations: Sample Size Estimation and Research Design Considerations - Page 4 preview image

Solution2.CompareandContrastRandomSelectionandRandomAssignmentRandom SelectionRandomAssignmentTherandomselectiondescribeshowtheresearch could draw the sample from thewhole population related tothe researchstudy.The random assignment describes about theprocessofresearchersthattheyusetoassign and draw the sample from differentgroups of the population (Trochim, 2006).Anexample.Ifaresearcherdraws50people from a group of 500 people with thehelp of random sample, then it would callrandom sampling.Ifthe researcher selects 50 people for newtreatment instead of remaining 450 peoplefromwholepopulationthenitwouldbecalledrandomassignment.Insomesituation, when the researcher does not usethe random sample to select the populationbut selectedthe some people for providingthe different treatment, then the randomselection could not be applied but randomassignment could be used (McNabb, 2010).Random selection depends over samplingprocessanditisdirectlyrelatedtotheexternalvalidityofthestudyresults(Trochim, 2006).Random assignment is related to the studydesignandalsorelatedtotheinternalvalidity of the study’s result.Solution3.Sample Size and Likelihood of a Statistically SignificanceIn the sense of statistics, significance is also a statistical term, which helps toensurethe difference and relationship takes place in two groups. It also depends on thesample size (StatPac Inc., 2013). For example, in a study, 50 people are selectedfor a test and the merit would be decided on the basis of male and female category.The males score 100 and the females score 98. The t-test gives the difference insignificance is .001. At the same time, there is no huge difference in 100 and 98.This result describes that there is small difference between the groups that couldnot make negative impact on the outcomes of the study.Therefore, it could be stated that the sample size makes the drastic impacts on theoutcomes of the study, whether the study is conducted for any population or for

Preview Mode

This document has 11 pages. Sign in to access the full document!

Report

Learning Tools

Writing Tools

Browse Resources

Statistical Analysis and Power Calculations: Sample Size Estimation and Research Design Considerations - Page 1

Study Now!

Document Details

Related Documents

Statistics - Univariate Inferential Tests

Statistics - Sampling

Statistics - Probability

Statistics - Principles of Testing

Statistics - Numerical Measures

Statistics - Introduction to Statistics

Statistics - Graphic Displays

Statistics - Cummulative Reviews

Statistics - Common Mistakes and Tables

Statistics - Bivariate Relationships

Company

Explore

Study Tools