Structure vs Unstructured communication the degree of standardization used with the data collection Structure instrument o In a highly structured questionnaire the questions and responses are completely standardized Same question same answers to choose from Aka fixed alternative question Open ended question closed ended a question in which the responses Fixed alternative question are limited to stated alternatives reply using their own words and are not limited to fixed set of possible answers type of question for which respondents are free to High structured questions are easy time consuming saves money and do not have to interpret biases o Consistency of responses answers However may force an answer when the person has no opinion Misleading High structure is most useful when the possible replies are well known limited in number and clear cut good for obtaining factual info Disguised vs undisguised communication Disguise disguised questionnaire attempts to hide the purpose or the sponsor of the study Undisguised makes the purpose of the research obvious 2 reasons to use disguised survey o When knowledge of the sponsor or topic of the survey is likely to cause respondents to change their answers o To create a more natural environment to collect data The ethics of disguise info To use of disguise can be seen as a way to deceive respondent or withhold Researchers believe the benefits of true usable info outweigh costs of the deception Debriefing after data have been collected using disguise the process of providing appropriate information to respondents Methods of Administering Questionnaires The main methods are personal interviews telephone interviews mail questionnaires and internet based questionnaires Three key aspects to consider o Sampling control refers to the ability of a particular method to identify and obtain responses form a sample of respondents from the target population o Information control is concerned with the number and types of questions that can be used and the degree to which researchers and or respondents might introduce error into the answers or their interpretation monetary costs of the different approaches resource issues such as the time and o Administrative control Personal Interviews Personal Interviews direct face to face conversation between an interviewer and the respondent o Can be a questionnaire o Can e conducted anywhere Mall intercepts a method of data collection in which interviewers in a shopping mall stop or interrupt a sample of those passing by to ask them if they would be willing to participate in a research study o Sampling control more sample control than other approaches Interviewer can verify identity of respondent Response rates are higher o Information control answers Ability to explain questions and respondents explain their Can see respondent reaction with product or pictures Gather the most info bc personal interaction HOWEVER errors Bias of the researcher Can write answer down wrong o Administrative control Most costly and time consuming Telephone interview telephone conversation between an interviewer and a Telephone interviews respondent o Sampling control Telephone samples do not include those without phones unlisted numbers or move to area after phone book is published and some numbers in book are disconnected Landline phones have given way This all increasing challenge of sampling control Random digit dialing telephone interviews in which the numbers to be called are randomly generated technique used in studies using Lately responses have not been good due to caller ID and answering machines o Information control Phone interviews use computer assisted interviewing using computers to manage the sequence of question and to record the answers electronically through the use of a keyboard Trust is built less likely of bias Should be between 5 10 min o Administrative controls Good response rates Data collected in days rather than months Supervisor listening in for quality control Mail Questionnaires A questionnaire administered by mail to designated respondents with an accompanying cover letter The respondents return the questionnaire by mail to the research organization o Sampling control Need accurate mailing list Get a list by using their contact list from previous customers or can purchase a mailing list Downside little control in getting a response from population o Information control No interviewer bias Can provide pics graphics or artwork Truly anonymous Disadvantages Cant explain questions Write out answers if open ended tedious for people Can view whole survey before starting o Administrative control Cost efficient Less supervision from management More time consuming Internet based Questions Questionnaire that relies on the Internet for recruitment and or completion two forms include e mail surveys and questionnaires completed on the web o Sampling control Have to give incentives for people to participate in survey online Some people are not using online sources like email Purchased email lists low response rates o Information control No interviewer bias Can provide graphics No ability to explain questions probe o Administrative control Inexpensive and quick turn around Chapter 7 Scales of Measurement Measurement of attributes rules for assigning numbers to objects to represent quantities o Measure attributes of objects and not object itself Example don t measure a person measure their income o Also the definition is broad it doesn t specify how the numbers will attitudes etc be assigned Different attributes have different qualities that dictate the rules for how numbers can be assigned Ex can t have gender on a 1 to 5 scale o 4 types of scales we use to do measurements nominal ordinal interval and ratio Nominal Scale Nominal scale classes of objects solely for the purpose of identification measurement in which numbers are assigned to objects or o Numbers used for identification and categorization Example circle 1 if female and 2 if male the numbers are assigned by the researcher and the researcher must be aware of what kinds of analysis are possible with different scales of measurement Ordinal scale basis of some order more than greater than of the objects measurement in which numbers are assigned to data on the o Example you could have 4 sodas and rank them from 1 favorite to 4 to least favorite Ordinal Scale Interval scale measurement in which the assigned numbers legitimately Interval scale allow the comparison
View Full Document
Unlocking...