Communication Inquiry COM 2010
Popular in Course
Popular in Communication, School Of
This 11 page Class Notes was uploaded by Mrs. Shania Schoen on Wednesday September 30, 2015. The Class Notes belongs to COM 2010 at Western Michigan University taught by Autumn Edwards in Fall. Since its upload, it has received 9 views. For similar materials see /class/216987/com-2010-western-michigan-university in Communication, School Of at Western Michigan University.
Reviews for Communication Inquiry
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/30/15
FINAL EXAMINATION Study Guide COM 2010 Wednesday April 27 800am 139 hm IiT til an zil jln39n i l all It lllf The final is worth I50 course points There will be 75 multiple choice and truefalse questions worth 2 points each The exam is comprehensive which means that material from units I 3 plus the new material on ethics is fair game The final does NOT include mathematical computations so you will not need formulas a calculator or extra paper However there will be conceptual questions about some of the mathematical procedures For instance even though you will not compute a standard deviation you need to know what it is and what it tells you Bring a PENCIL and arrive promptly From Unit I Know understand and be able to apply the following r Operation 9 Theoretical Problem 9 Research question or hypothesis 9 Definition of variables 9 Methodology quantitative or qualitative 9 Data collection 9 Data Analysis Independent Variables Predict the outcomes state amp in a hypothesis or research question causes Deendent Variables Deend on indeendent effects J x i 7 Bill orquot IIII I Title Abstract Introduction Purpose Statement Independent Variable Dependent variable Context Justification Review of literature Theoretical Perspective Informational review Conceptual definitions Argument RQ s or Hypothesis Research Questions Descriptive Relational Hypothesis One Tailed higherlower directional difference Two Tailed non directional predictsrelationship Population All members of a group eg Men in the US Sample A subgroup of the population eg A portion of men in the US Parameter Fact about the population eg Average age of men in the US Statistic Fact about the sample eg Average age of men in the sample nferential Statistic Makes predictions about the population Descri tive Statistic Describes the sam le Measurement scale Method of measuring or operationalizing variables 4 general types Nominal Ordinal Interval 4 kinds Ratio Nominal Categories groups classes eg Sex ethnicity religion must be mutually exclusive equivalent and exhaust Ordinal Ranked ordered groups each group listed has more of somethini than the one before ec socio economic class yli f cw l 12 iiill mn Strict C39llrllilquot 1ltltllquotirliiliign w cases in a samp e are different from one another 39 r9 3 ways ho Only apply to intervalratio data l 2 Range Highest score subtracted from lowest score Standard Deviation Measures the average distance between each score and the mean score Variance Measures the total variability of a sample variance is symbolized by S2 519 ll quot7 0 1a l 4 Simple Random Sampling This is the ideal choice as it is a perfect random method Using this method individuals are randomly selected from a list of the population and every single individual has an equal chance of selection Systematic Sampling Systematic sampling is a frequently used variant of simple random sampling When performing systematic sampling every kth element from the list is selected this is referred to as the sample interval from a randomly selected starting point For example if we have a listed population of 6000 members and wish to draw a sample of 2000 we would select every 30th 6000 divided by 200 person from the list In practice we would randomly select a number between i and 30 to act as our starting point Stratified Sampling Stratified sampling is a variant on simple random and systematic methods and is used when there are a number of distinct subgroups within each of which it is required that there is full representation A stratified sample is constructed by classifying the population in sub populations or strata base on some well lltnown characteristics of the population such as age gender or socio economic status The selection of elements is then made separately from within each strata usually by random or systematic sampling methods Stratified sampling methods also come in two types proportionate and disproportionate n proportionate sampling the strata sample sizes are made proportional to the strata population sizesFor example if the first strata is made up of males then as there are around 50 of males in the UK population the male strata will need to represent around 50 of the total sample In disproportionate methods the strata are not sampled according to the population sizes but higher proportions are selected from some groups and not others This technique is typically used in a number of distinct situations Cluster or Multi stage Sampling Cluster sampling is a frequently used and usually more practical random sampling method It is particularly useful in situations forwhich no list of the elements within a population is available and therefore cannot be selected directly As this form of sampling is conducted by randomly selecting subgroups of the population possibly in several stages it should produce results equivalent to a simple random sample The sample is generally done by first sampling at the higher levels eg randomly sampled countries then sampling from subsequent levels in turn eg within the selected countries sample counties then within these postcodes the within these households until the final stage is reached at which point the sampling is done in a simple random manner eg sampling people within the selected households The levels in question are defined by subgroups into which it is appropriate to subdivide your population Cluster samples are generally used if No list of the population exists Well defined clusters which will often be geographic areas exist A reasonable esTimaTe of The number of elemenTs in each level of clusTering can be made OfTen The ToTal sample size musT be fairly large To enable clusTer sampling To be used effecTively drill STandard Error of The Mean SEM AnoTher word for margin error 95 and 98 which of The Two has The biggesT range From UniT ll Know undersfond and be able To apply The following rlz l a i u Research quesTions RQs uesTions abouT relaTionships among variables or differences beTween groups CharacTerisTics of RQs Clearly sTaTes in form of a quesTion Eg Does pracTice significanle decrease public speaking anxieTy2 includes 2 or more variables TesTable HypoTheses EXpecTaTions based on The assumed relaTionship beTween variables PredicTions used To TesT Theories CharacTerisTics of hypoTheses Clearly sTaTed in The form of a declaraTion includes 2 or more variables TesTable consisTenT wiTh previous research Research HypoTheses H1 predicTs differences beTween groups Non direcTional STaTes ThaT a difference beTween groups exisT buT doesn T specify The naTure of The difference Two Tailed H1 2m u2 TranslaTion The research hypoThesis is ThaT The populaTion average of group one is noT equal To The populaTion average of group Two DirecTional sTaTes ThaT a difference beTween groups exisTs and specifies The naTure of The difference one Tailed Null HypoTheses Ho PredicTs no difference beTween groups Non DirecTional sTaTes ThaT There is no difference beTween The group DirecTional opposes The difference beTween groups sTaTed in The research h oThesis i v i a 3 in l n inTerpreTaTIon discussion and desIgn we Talllt abouT The research hypoThesis in sTaTisTical analysis we Talllt abouT null hypoThesis f The sTaTisTical analysis indicaTes ThaT There is a significanT difference beTween groups say I rejecT The null hypoThesis The besT we can do is rejecT The null hypoThesis Our sToTisTics con T prove Tth The resedrch hypoThesis is definiTely True f The sTdTisTicol dndlysis indicoTes Tth There is no difference beTween rou s so I rerin The null hypoThesis Type 1 error dlphd is The chdnce of mdllting 0 Type one error or Tolser rejecTing The null Alphd moy dlso be colled rejecTion region significonce level p for probobiliTy Alphd is usudlly seT dy 05 or less so There is d 5 chdnce Tth we misleltenl rejecTed 0 True null 391 leaf independenT sdmple T TesT TesTs for d sTdTisTicolly significonT differenT beTween The medns of Two independenT groups HypoThesis sTudenTs who hove lelten 0 speech cldss will be beTTer spedllters Thon sTudenTs who hove noT lelten 0 speech cldss H pcloss gtpno cldss H1 pcloss lt pno cldss Two groups Group lstudenTs who Toollt 0 speech cldss Group 2 sTudenTs who did noT Take 0 speech cldss Procedure Rdndomly selecT lO sTudenTs who hove hdd 0 speech cldss ond lO sTudenTs who hove noT Eoch sTudenT gives 0 speech in fronT of on experT judge who gives The speech on effecTiveness score beTween l ond lO SsTeTi vorlone True group differences differences beTween groups Error vorionce chonce differences differences wiThin groups Why mighT some people why hdd 0 speech cldss give 0 bod speech Why mighf some people who hdd no speech cldss give 0 good speech Sysfemdfic and Error Voridnce Problem we need 0 woy fo compore error vorionce fo sysfemdfic voridnce Solufion The independenf sdmple f fesf If works by compdring sysfemdfic voridnce fo error voridnce EJUl39l nl will u if o Assumpfions of fhe f fesf o l Independenf rdndom sdmples Independenf sdmples ore nof reldfed Rondom dll coses hove on equdl chdnce of being selecfed o Infervol or Rdfio DV 0 Normol Disfribufion o Homogeneify of Voridnce The dmounf of vorionce in one group is simildr fo fhe dmounf of vorionce in fhe ofher group 0 So fhe f fesf is PARAMETRIC If mdlltes dssumpfions obouf fhe chorocferisfics of fhe populdfion fhe pordmefers o Formuld o flnl n2 2 Xi X2sqrsl2 o Sfeps conf 0 Find crificol f on ffoble f fhe f colculofed in sfep 4 is gt fhe crificol f Rejecf fhe Null f fhe f colculofed in sfep 4 is lt fhe crificol f Refdin fhe Null 0 6 Reporf fhe resulfs Indicofe whefher fhe groups differed significonfly how fhey differed and five fhe medns ond sfdnddrd deviofion of edch group ds well ds fhe f fesf informdfion o If you REJECT Students who have taken a speech cass M SD are significantly better speallters than students who have not taken a speech cass M SD t df plt 05 o If you RETAIN There is no significant difference between students who have taken a speech cass M SD and students who have not taken a speech cass M SD t df pgt 05 From Unit Ill Know understand and be able to apply the following Reiabiity Reiabiity consistency Is the measure stable Does the measure produce the same measurement over time and across situation Assessing Reiabiity How do you determine whether a measure is reliable Compute a reliability coefficient Correlation Relationship Can range from 0 no reliability to 1 perfect reliability 80 and above is considered good 70 is considered adequateacceptable Anything below 60 is considered unacceptable Researchers are expected to produce evidence that their measures are reliable for journal articles Types of reliability Test retest How a measure performs across Time Give The measure Twice and see if There is consisTency beTween scores SpliT Half Measures across parTs of The scale Divide The measure inTo 2 parTs and see if There is consisTency beTween Those parTs nTernal consisTency Measures across all iTems on The scale CompuTe a correlaTion beTween each iTem and all oTher iTems on The scale nTercoder ReliabiliTy Measures agreemenT beTween 2 or more coders Checllting To see if people raTe Things The same way ReliabiliTy vs validiTy You can have reliable measure ThaT is noT valid BuT you can T have a valid measure ThaT is noT reliable n oTher words reliabiliTy is a precondiTion of validiTy buT does noT guaranTee validiTy Types of validiTy Face validiTy Researcher loollts aT The conTenT of The measure To deTermine wheTher iT appears valid CriTerion validiTy Examine validiTy of a measure by comparing iT To some ouTside criTeria ConvergenT how does iT compare To oTher measures of The same Thing PredicTive does The measure predicT performance ConsTrucT validiTy Researcher examines The measure in relaTion To oTher TheoreTically relaTed measure Research proces where reliabiliTy and validiTy fiT in C lir39 Elli MeasuremenT 0 Variables Scales Survey insTrumenTs RaTing procedures Coding proTocols ii l Good Measures haveReliabiliTy ValidiTy ReliabiliTy ReliabiliTy consisTency Is The measure sTable Does The measure produce The same measuremenT over Time and across siTuaTion Example Assuming yourweighT does noT change A reliable scale gives you The same number every Time An unreliable scale gives you a differenT number every Time Jll c U ill WhaT can cause a measure To be unreliable FamiliariTy wiTh The measure formaT FaTigue EmoTional STrain EnvironmenTal FacTors Physical FacTors of The measure Tallter FlucTuaTion in Memory Prior pracTiceexperienceknowledge wiTh The variabiliTy being measured A research meThod which quesTioning sTraTegies are used To gaTher descri Tive informaTion abouT a o ulaTion Large random represenTaTive For small populaTions samples should include mosT of The populaTion ExploraTion WhaT variables should be sTudied in This populaTion DescripTion WhaT are The characTerisTics variables of a populaTion2 WhaT are The relaTionships among The characTerisTics variables of a o ulaTion Response raTe is The percenTage of people who compleTe a survey 60 is minimum 75 and above is excellenT To increase response raTe Provide incenTives Cash or prizes Moral MulTile oorTuniTies mailins hone calls W n i 4 Closed can be answered in very few words Open inviTe long deTailed varied responses DirecT asks for desired informaTion ouTrighT ndirecT uses back door approach To geT desired informaTion General is vague Terms aren T really defined Specific is clear deTailed and precise Questionnoires stdndordized written questions Telephone Stdndordized ordl questions Focus groups group interviews che to fdce interview conversdtiondl formot From Ethics End of WWII 23 Ndzi doctors ond scientists were tried for murder of concentrdtion comp inmdtes who were used ds resedrch subjects All but 8 were sentenced to prison or deoth According to the code Get informed consent Bose resedrch on prior dnimdl worllt Risks should be justified by onticipdted benefits Only qudlified scientists must conduct resedrch Physicol ond mentdl suffering must be ovoided Resedrch in which deoth or disdbling injury is expected should not be conducted Respect for persons treot individudls ds outonomous humdn beings Informed consent Respect privocy Beneficience minimize horms ond moximize benefits Use best possible resedrch design Reseorchers must be qudlified to perform procedures ond hdndle the risks Justice Tredt people fdirly ond design resedrch so thdt burdens ond benefits ore shored equitdbly Requirement to select subjects equitdbly Avoid exploitotion of vulneroble populdtions How know was on experiment Tredtment independent variable Environment dependent variable Behdvior Control Any effects see in terms of dependent voridble ore bdsed on independent voridble
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'