Graphic Era University, Dehradun POTENTIAL SOURCES OF ERROR Several potential sources of error can affect a research design

. A good research design attempts to control the various sources of error. Although these errors are discussed in great detail in subsequent chapters, it is pertinent at this stage to give brief descriptions. The total error is the variation between the true mean value in the population of the variable of interest and the observed mean value obtained in the marketing research project. For example, the average annual income of the target population is Rs70,000, as determined from latest census records, but the marketing research project estimates it as Rs62,000 based on a sample survey. Total error is composed of random sampling error and nonsampling error.
Total Error Random Sampling Error Response Error Researcher Error
Surrogate Information Error Measurement Error Population Definition Error Sampling Frame Error Data Analysis Error

Non-sampling Error Non-response Error Respondent Error
Inability Error Unwillingness Error

Interviewer Error
Respondent Selection Error Questioning Error Recording Error Cheating Error

Random Sampling Error Random sampling error occurs because the particular sample selected is an imperfect representation of the population of interest. Random sampling error is the variation between the true mean value for the population and the true mean value for the original sample. For example, the average annual income of the target population is Rs70,000, but it is only Rs69000for the original sample, as determined from the mail panel that are believed to be accurate Nonsampling Error Nonsampling error can be attributed to sources other than sampling, and they may be random or nonrandom. They result from a variety of reasons, including errors in problem definition, approach, scales, questionnaire design, interviewing methods, and data prep and analysis. For example, the researcher designs a poor questionnaire, which co several questions that lead the respondents to give biased answers. Nonsampling consists of 1) Non response errors 2) Response errors.

Prepared By-Sanjeev Kumar (Faculty Management))


questioning. For example. recording. sampling frame. and new numbers in service. a) Surrogate information error may be defined as the variation between information needed for the marketing research problem and the information sought b researcher. d) Sampling frame error may be defined as the variation between the population defined by the researcher and the population as implied by the sampling frame (list) used. Non response error arises when some of the respondents included in the sample do not respond. or respondents. Nonresponse will cause the net or resulting sample different in size or composition from the original sample. and data analysis errors. interviewers. For example. e)Data analysis error encompasses errors that occur when raw data from questionnaires are transformed into research findings. Response error is defined as the variation between the true mean value of the variable in the net sample and the observed mean obtained in the marketing research project. Prepared By-Sanjeev Kumar (Faculty Management)) 2 . Seeking to measure consumer preferences. the average annual income' Rs69. Nonresponse error is defined the variation between the true mean value of the variable in the original sample and true mean value in the net sample. Response errors made by the interviewer include respondent selection . the average annual income is Rs 70000 for the original sample.000in the marketing research project Response errors can be made by researchers. but Rs69000 for the net sample. measurement population definition. the researcher employs a scale that measures perceptions rather than preferences. The primary causes of non response are refusals not-athomes. For example. a nonreader is selected for the interview but classified as a reader of the Wall Street Journal in the 15 to 19 years category in order to meet a difficult quota requirement. For example. c) Population definition error may be defined as the variation between the actu population relevant to the problem at hand and the population as defined by the researcher.Graphic Era University. Dehradun 1) Nonresponse Error. For example.Errors made by the researcher include surrogate information.000 for the net sample. but is estimated as Rs62. an inappropriate statistical procedure is used resulting in incorrect interpretation and findings. the researcher obtains information consumer preferences because the choice process cannot be easily observed. 2) Response error arises when respondents give inaccurate answer their answers are misreported or misanalyzed. For example. and cheating errors. the telephone directory used to generate a list of telephone numbers does not accurately represent the population of potential consumers due to unlisted. a) Respondent selection error occurs when interviewers select respondents other than those specified by the sampling design or in a manner inconsistent with th~ sampling design. in a readership survey. disconnected. instead of obtaining information on consumer choice of a new brand (needed for the marketing research problem). b) Measurement error may be defined as the variation between the information sou~ and information generated by the measurement process employed by the researcher.

Dehradun b) Questioning error denotes errors made in asking questions of the respondents or in not probing when more information is needed. a respondent indicates a neutral response (undecided) but the interviewer misinterprets that to mean a positive response (would buy the new brand). suppose a mail survey is being conducted to determine consumer preferences for purchasing fashion clothing from department stores. an interviewer does not use the exact wording given in the questionnaire. d) Cheating error arises when the interviewer fabricates answers to a part or all of the interview. a respondent intentionally misreports reading Time magazine in order to impress the interviewer. nonrespondents are likely to differ from respondents in terms of salient variables. Hence. For example. an interviewer does not ask the sensitive questions related to respondent's debt but later fills in the answers based on personal assessment. the selection of a large sample size does not allow for follow-up mailings. whereas many forms of nonsampling error defy estimation. the two follow-up mailings will more than offset this loss by decreasing nonresponse error Prepared By-Sanjeev Kumar (Faculty Management)) 3 . Sampling error can be calculated. and other factors. researchers deliberately increase a particular type of error to decrease the total error by reducing other errors. Respondents may intentionally misreport their answers because of a desire to provide socially acceptable answers. question content. whereas random sampling error is relatively small in magnitude. Although decreasing the sample size will increase random sampling error. For example. boredom. avoid embarrassment. fatigue. interpreting. Nonsampling error is likely to be more problematic than sampling error. nonsampling error has been found to be the major contributor to total error. A large sample size has been selected to reduce sampling error.Graphic Era University. However. faulty recall. while asking questions. For example. A particular type of error is important only in that it contributes to total error. For example. it may be desirable to reduce the sample size to make money available for follow-up mailings. Moreover. b)Unwillingness error arises from the respondent's unwillingness to provide accurate information. or please the interviewer. question format. A response rate of 30 percent may be expected. a respondent cannot recall the brand of yogurt purchased four weeks ago. past experience indicates that the response rate could be increased to 45 percent with one follow-up and to 55 percent with two follow-up mailings. Sometimes. Given the limited budget for the project. For example. Given the subject of the survey.The point is that total error is important. For example. c) Recording error arises due to errors in hearing. and recording the answers given by the respondents. a) Inability error results from the respondent's inability to provide accurate answers. Respondents may provide inaccurate answers because of unfamiliarity. Response errors made by the respondent are comprised of inability and unwillingness errors.

Sign up to vote on this title
UsefulNot useful