ADV 281 Research Methods Test 4 About surveys provide point in time snapshot four data collection methods personal face to face telephone mail and self administered online Personal Interviews Intercept recruit Rs on the spot location dependent Pre recruited contact in advance pre screening set appointment Tend to provide highest level of data quality dut to Personal administration of interview by trained interviewer Face to face contact Other advantages include Work well for complex questionaires Permit use of visual stimuli Can confirm respondent understanding But expensive and time consuming Telephone interviews Conducted from central location use CATI computer assisted telephone interviewing Advantages Easily administer complex skip patterns lower cost can obtain true random sample Disadvantages Short length simple questions hard to get sensitive data high refusal rates no visual stimuli Mail surveys Respondents receive packets cover letter questionnaire instructions for completion and return return envelope incentrive optional Randomly selected sample of population Mail panel versus randomly selected sample higher response rates greater cost efficiencies more efficient pre screening mail surveys pro cost efficiency respondent convenience con lower response rate limited online surveys advantages can automatically execute complex skip patterns random assignment survey can be experiment lower cost world wide sample disadvantages sample representativeness due to differences in internet access Criteria for selecting data collection method cost length response rate geo Coverage sample characteristics time telephone online fastest personal interviews slowest sample who responds interview circumstances for respondents and administrative monitor interviewer quality control information needs phone and mail elicit more honest interview more thorough topic and Q complexity interview length Complexity personal interviews best when both topics and questionnaire are complex telephone works well when complexity is entirely due to questionnaire structure online appropriate for relatively simple topics with or without questionnaire complexity mail appropriate for simple topics and simple questionnaires Response Rate Percentage of valid sample who provide a valid completed interview Calculated by diving number of respondents completing survey by total number of respondents in the valid sample Response Rate example Sources and remedies for nonresponse error Source refusal not at homes Remedies advance notification incentives callbacks recontact Note Text is specific about how these remedies may be used for personal telephone mail and online Research questions Free samples on purchase rate Measurement The way by which abstract typically not observable concepts are linked to observable events Rules for assigning numbers to objects in such a way as to represent qualities of attributes Measuring concepts Scales Criteria for good measurement Measurement process Stage one identify and define concept of interest Identify concept of interest develop conceptual definition develop operational definition Stage two specify an observable event Determine level of measurement select question type Stage three evaluate and reuse observable event Rewrite question access validity and reliability Concept Generalized idea about class of objects attributes occurrences or processes Conceptualize conceptual definition Verbal explanation of meaning of a concept expresses central or core idea of the concept defines what the concept is and what it is not Commonly access attitudes toward products ads and brands What is attitude Psychological tendency to evaluate something with some degree of favor or disfavor Attitude towards an ad Predisposition to respond in a favorable or unfavorable manner to a particular advertising stimulus during a particular exposure situation Akers brand personality Set of human characteristics associated with a brand Sincerity excitement competence sophistication ruggedness Operation ism operation definition Concepts in scientific theories must be grounded in linked to observable event that can be measured An explanation that gives meaning to a concept by specifying the activities or operations necessary to measure it Operational definition removed concept from the feelings intuitions of a particular individual and allows it to be tested by anyone who can carry out the measurable operations From conceptual to operational definition Explicitly specify the concept of interest Explore different aspects of the concepts meaning Explicitly state what can be observed Evaluate and select one or more of the alternatives Example operational definition of attitude toward the ad Measuring anything that exists ie love Levels of measurement Different levels of measurement provide different types of information Nominal Numbers or letters assigned to the object served as labels for identification or classification Goal Categorize responses through assignment to mutually exclusive and exhaustive categories Dichotomous yes or no Checklist Combines a series of related dichotomous into single question Less time consuming and tedious Must define criteria on which checklist items will be selected Must be constructed I reduce bias order equal number of pos neg items Example What words describe how you felt about the commercial you just saw Nominal scale Ordinal Arranges objects or alternatives according to their magnitude in an ordered relationship Goal Order responses according to a predefined characteristic Rank Order Scaling Interval Arranges objects according to their magnitudes and also distinguishes in units of equal magnitude Goal Provide an estimate of the relative distance between items Graphic semantic differential likert Ratio Has absolute rather than relative quantities and an absolute zero where there is the absence of an attribute Periods of time dollars spent number of items purchased Constant sum scale Selecting a level of measurement Determined by Characteristic or attribute being measured Desired depth of detail Level of detail required for data analysis General rule collect data at highest level of measurement Leles of measurement determines what information you will have about the object of study determines what you can and cannot say about the object Level of measurement determines what type of stat analysis you can perform high level scales permit more sophisticated analyses Gorillas Simons and