Total error in marketing research is the deviation between the true value and the observed value in the project. Numerous errors can affect the quality and credibility of the research findings. Two main types of errors have been identified: Random sampling errors and Non-random sampling errors. The first derives from how well the sample selected represent the population being studied while the latter represents all types of error that may occur from sources other than sampling.
Since non-sampling error is very broad it has been divided into response errors and non-response errors. While non-response errors are mainly cause by refusal to participate in a survey or not-at-home, response errors are generated by either the participant, or the researcher or the interviewer. Sampling error affect the total error however that effect is relatively small to the consequences of non-sampling error. Researcher should aim to design their research in a way to minimize the total error instead of a particular type of error in order to get the most accurate results.
In their book, -Marketing Research- Malhotra, Briks & Wills elaborated about sample sampling errors and non-sampling errors by describing them as follows.
“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 true mean value for the population and the true mean value for the original sample.” Therefore sampling error appears when the characteristic of a sample is representing the entire population being under research. For instance, if we are studying the average weight in Oman and we take a sample of ten thousand persons living in the Sultanate, their aver...
... middle of paper ...
...is needed. For example, while asking questions an interviewer does not use the exact wording or prompts as set out in the questionnaire.
Recording error arises due to errors in hearing, interpreting and recording the answers given by the participants. For example, a participant indicate a neutral response(undecided) but the interviewer misinterprets that to mean a positive response (would buy the new brand)
Works Cited
Malthotr, N , Birks, D & Wills, B (2012). Marketing Research AN APPLIED APPROACH. 4th ed. England: Pearson Education Limited. Pages 101-590.
Ministry of Education NewZealand Government. (2013). Glossary. Available: http://seniorsecondary.tki.org.nz/Mathematics-and-statistics/Glossary/Glossary-page-N. Last accessed 12/12/13.
Wikipedia. (2012). Non-sampling error. Available: http://en.wekipedia.org/wiki/non-sampling_error. Last accessed 12/12/13.
Two sampling methods include mail surveys and convenience sampling, a variation of a nonprobability sample. Mail surveys, inexpensive way to contact individuals over a large geographical area, provide anonymity to the respondent, and eliminate interview bias. Convenience sampling, a nonprobability sample, the only criteria is the convenience of the unit to the researcher, fast and uncomplicated, but the sampling error not determined.
Random representative sampling is a method of sampling that uses random selection to obtain its samples. By making sure that everybody has an equal chance at being selected, random representative sampling ensures diverse samples. Using the example in paragraph one, a random representative sample allows you find the statistics on all the company’s employees without interviewing all them. Random representative sampling is important for getting accurate poll results because it allows you to find the view of a population while making sure that the poll is not biased in any way.
In statistics, a population is a collection of individuals, things, events, etc. The population is the topic that one wants to make inferences on, whereas a sample is a subset of the population that is being collected—to be studied. After the sample is studied in statistics, one draws an inference of the population. There are four general sampling methods used in statistics: representative sample, random sample and quasi-random sample, stratified and quota sample, convenience sample, and purposive sample. A representative sample should be unbiased and thus properly indicate a characteristic of the entire population. In a random sample nothing is biased; in other words, every individual, thing or event in the population has the same chance of being selected for the sample. Therefore, because of the randomness of the sampling, the selection of one item from the population in no way effects the selection of another item. A quasi-random sample is simply a number (nth), which is
Armstrong, Gary, and Philip Kotler. Marketing: an introduction. 11th ed. Upper Saddle River, NJ: Pearson Prentice Hall, 2013. Print.
Another challenge would be the ability to generalize the results. Actually, it is basically an assumption that the samples reflect the views of the larger market. Though, this has been minimized through the use of systematic sampling.
The articles, published after 1996, contain varied methods of research attainment, but share similarities such as being a self-survey, having a small sample size, and being
Etzel, Michael J., Stanton, Bruce J., Stanton, William J. (2004). Marketing. (13th ed.). Boston: McGraw-Hill.
Armstrong, G, Adam, S, Denize, S, Kotler, P, 2010, Principles of Marketing 5th Edition, Pearson Australia Group, Frenchs Forest
Armstrong G. & Kotler P. (2007) Marketing: An Introduction 8E Upper Saddle River, NJ Pearson Prentice Hall Publishers
Kotler, J., & Keller, K. (2012). A framework for marketing management. Essex: Pearson Education Ltd.
...sis, which could make all results invalid. With any research projects, limitations will be present. It is important to attempt to eliminate some of these causes in order to complete a thorough, accurate study. In future projects, this study could go about researching the issue in different ways. Perhaps using a larger sample size would be conducive for accurate results. A larger sample size helps reduce and even out any possible errors caused by those who do not answer truthfully. Also, keeping the surveys mainly anonymous would help to receive more truthful and accurate responses from participants. Participants may be fearful of judgments on open interviews or phone interviews, which could affect responses. In order to obtain as accurate results as possible, a future study would need to find ways to survey participants in a confidential way that feels comfortable.
Malhotra, N.K., 2002. Basic Marketing Research: A Decision-Making Approach. Upper Saddle River, New Jersey.: Prentice Hall.
Lack of response is the main disadvantage for mail surveys. The group survey is another low cost form, however the individual respondent is interviewed in a group. The disadvantage with group surveys are the logistics of marshaling the respondents to one location and the perception by respondents that grouping posses less anonymity. Electronic surveys are a relatively new addition in survey research and could very well become comparable to the telephone survey. Electronic surveys are advantageous for the low cost as well as ease in delivery. Because the delivery method is through internet, and the general population does not
Grover, R & Vriens, M 2006, The handbook of marketing research: Uses, misuses, and future
This paper discusses different types of sampling techniques used in quantitative research. It begins by looking at probability sampling (also known as random sampling) before discussing non-probability sampling (non-random sampling). The discussion ends by looking considerations that should be made before selecting a sampling technique before concluding. Because quantitative researchers prefer probability sampling and only use non-probability on rare occasions the e...