Click the card to flip 👆
1 / 91
Terms in this set (91)
Computer-Administered Surveys Advantages¡Many user-friendly features ¡Highly efficient ¡Reduction of interview evaluation concern in respondentsComputer-Administered Surveys Disadvantages¡Requires computer-literate and Internet-connected respondents ¡Respondent misrepresentationMixed-Mode Surveys Advantages¡Multiple advantages to achieve data collection goal ¡Could use online survey to quickly reach portion of population with Internet access and telephone calling to reach those without Internet access.Mixed-Mode Surveys Disadvantages¡Survey mode may affect response ¡Data complexityWays to Gather DataIn-home Mall-intercept In-office Telephone Fully Automated Online Group self-administered Drop-off survey Mail surveyIssues with mail surveysNonresponse Self-selection biasCentral Location Telephone Surveying•Good and efficient quality control •Interviewing process can be monitored •Control over interviewer schedulesPanel Company definitionan institution that recruits potential survey respondents who participate for compensationPanel company advantages•Fast turnaround •High quality •Database information •Access to targeted respondents •Integrated featuresPanel company disadvantages•Not random samples •Overused respondents •CostMeasurement definition•determining a description or amount of some property of an object that is of interestProperties definition•specific features or characteristics of an object that can be used to distinguish it from another objectObjective properties•are observable and tangible.subjective properties•are unobservable and intangible and must be translated onto a rating scale via scale development.3 types of measurementsnominal ordinal and scaleNominal scale measurements•scales label objects. EX male or femaleOrdinal scale measurements•indicate only relative size differences between objects. Allow to rank order the responses. EX please rank candy pref by 1-5Scale measurements•those in which the distance between each level is known EX rank familiarity with shows from 1-20Ratio Scale measurementsEX please provide age or Approximately how many times in the last month have you purchased something over $10 Or How much do you think a typical purchaser of a $250,000 term life insurance policy pays per year for that policy?Likert ScaleFoscuses on good traits of product ex indicate degree to which you agree with following from 1-10Semantic Differential Scale•A series of bipolar adjectives for the various properties of the object under study •A good way to measure a brand, company, or store image Halo effectStapel Scalerelies on positive and negative numbers, typically ranging from +5 to -5Slider ScaleCan move figure to portray agreement with statementCommonly Used Scales for ConstructsBrand/store image Frequency of use Importance Intention to purchase Lifestyle or opinion Performance or attitude SatisfactionReliability•respondent responds in the same or a similar manner to an identical or nearly identical measure. A reliable measure is consistentValidity•accuracy or exactness of the measurement. A valid measure is truthful.The Questionnaire: Functions•Translates the research objectives into specific questions asked of respondents •Standardizes those questions and the response categories so every participant responds to identical stimuli •Serves as an enduring record of the research •Depending on the data collection mode used, such as online, a questionnaire can speed up the process of data analysis •Depending on the data collection mode used, such as online, a questionnaire can speed up the process of data analysisQuestionnaire Design & Development•Systematic process •Aim for understandable, unambiguous, and unbiased •Should minimize question biasQuestionnaire Guidelines DO•The question should be focused on a single issue or topic. •The question should be brief. •The question should be grammatically simple. The question should be crystal clearQuestionnaire Guidelines DONT•Do not "lead" the respondent to a particular answer. •Do not use "loaded" wording or phrasing. •Do not use a "double-barreled" question. •Do not use words that overstate the case.Leading questionsquestions that predispose a respondent to answer in a certain wayLoaded questionsThey are questions which contain emotive language which is likely to produce an emotional reaction in the respondent.double-barreled questionsquestions that attempt to get at multiple issues at once, and so tend to receive incomplete or confusing answers EX Were you satisfied with food and service?Overstated questionsplaces undue emphasis on some aspect of the topicQuestionnaire Organization•Well-organized questionnaires motivate respondents to be conscientious and complete. •Poorly organized questionnaires discourage and frustrate respondents and may even cause them to stop answering questions in the middle of the survey.Computer-Assisted Design•Software that allows users to use computer technology to develop and disseminate questionnaires and to retrieve and analyze data gathered by the questionnaire •Easy, fast, friendly, and flexible •Question types, question libraries, real-time data capture, and downloadable datasets Skip and display logicCoding•Numbers associated with question responses to facilitate data entry and analysis. •The codes for an "all that apply" question are set up as though each possible response was answered with "yes" or "no."Pretesting•The codes for an "all that apply" question are set up as though each possible response was answered with "yes" or "no."What is the group under the study?The populationConcepts in samplingPopulation Sample Sample Unit Sample Frame Sampling frame error Sampling errorReasons for samplingPractical considerations such as cost and population size Inability to analyze huge amounts of data generated by a censusProbability samplesones in which members of the population have a known chance (probability) of being selected into the sample. Only one that is truly representative.Non-probability samplesinstances in which the chances (probability) of selecting members from the population into the sample are unknownSimple random samplingthe probability of being selected into the sample is "known" and equal for all members of the populationprobability of selectionsample sample/population sizerandom device methodinvolves using a procedure or apparatus that assures that every member of the population has the same chance of being included in the sample. Can use a blind draw or random number A common practical application is Random digit dialing (R D D)Systematic samplingway to select a random sample from a directory or list Much more efficient than simple random sampling Uses a skip interval for selectionSkip intervalpopulation list size/sample size. EX every 10th person in the list.Cluster samplingmethod in which the population is divided into subgroups, called "clusters," each of which could represent the entire populationArea samplingForm of cluster sampling - the geographic area is divided into clusters.• One-step vs. two-step area samplingDisadvantage of Cluster SamplingCluster Specification ErrorCluster Specification Erroroccurs when clusters are not homogeneous.Stratified samplingSeparates the population into different subgroups and then samples all of these subgroups◦ Is used when working with a "skewed" population◦ May require the calculation of a "weighted mean"Nonprobability Sampling MethodsNot based on fairness, equity, or equal chance.Convenience samplingsamples drawn at the convenience of the interviewer (such as mall interviews)Chain referral (Snowball) samplesrequire respondents to provide the names of prospective respondentsPurposive samplesrequires a judgment or an "educated guess" as to who should represent the populationQuota samplesspecifies percentages of the total sample for various types of individuals to be interviewedOnline sampling techniquesOnline panel samples River samples E-mail list samplesOnline panel sampleslarge numbers of individuals who have agreed to participate in online surveysRiver samplescreated via the use of banners, pop-ups, or other online devices that invite website visitors to take part in the surveyEmail list samplespurchased or otherwise procured from someone or some company that has compiled email addresses of opt-in members of the population of interestSteps in a sample plan1. define problem 2. obtain a sample frame 3. decide on the sample method 4. decide on sample size 5. draw the sample 6. validate the samplemargin of sample error or simply sample errorA random sample will always have some inaccuracy, A random sample size can be a tiny percent of the population size and still have a small margin of sample error.The only perfectly accurate sample is acensusThe larger a random sample is, the more ____ it is, meaning the ____sample error it has.accurate;lessOnly a ________ sample (random sample) is truly representative of a populationprobability. The size of that random sample determines the sample accuracySample method determines__________ while Sample size determines__________representativeness; accuracySample accuracya measure of how closely it conveys the true values of the population it represents.Nonsampling errorpertains to all sources of error other than sample selection method and sample sizeSampling erorinvolves sample selection and sample sizeSample accuracy and representativenessrefers to how close a random sample's statistic is to the true population's value it representsVariability (p*q)amount of dissimilarity in respondents' answer to a questionWhich of these is more accurate? A large probability sample or a small probability samplelarge probability sampleSample size variablenZ variablestandard error associated with the chosen level of confidence (typically, 1.96)p variableestimated percent in the populationq variable100 − pe variableacceptable sample error expressed as a percentWays to estimate variabilityUse data from a previous study on the target population • Conduct a pilot study of the target population • Assume worst case maximum variability (p and q = 50% each)What are the 7 axioms?The only perfectly accurate sample is a census A random sample will always have some inaccuracy, which is referred to as "margin of sample error" or "simply sample error" The larger a random sample is, the more accurate it is, meaning the less sample error it has Margin of sample error can be calculated with a simple formula and expressed as a ±% number You can take any finding in the survey, replicate the survey with a random sample of the same size, and be "very likely" to find the same finding within the ±% range of the original sample's finding In almost all cases, the margin of sample error of a random sample is independent of the size of the population random sample size can be a tiny percent of the population size and still have a small margin of sample error