On July 24th,1824 *“The Harrisburg Pennsylvanian” *printed a report of a straw vote taken in Wilmington, Delaware. This vote showed some of the early methods used in market research. Although, this event was not market research in itself, it used some of the early methods of market research. However, it was not until 1901 that evidence of market research became frequent enough to indicate that a new business field had made a prominent start. The need for knowledge had several businessmen looking for theories and methods to better understand the market, which ultimately created the market research industry we know today. Market research encompasses numerous methodologies, but the most common one is survey research.

In order to conduct any type of research, it is necessary to have a quality sample of the targeted audience. Reliable sample is pivotal in market research. One of the key factors involving sample is selecting the correct sample size. This is crucial when conducting a study, if a sample size is too big this will lead to a waste of resources. Consequently, not having enough sample will lead to an inaccurate representation of a population. Survey accuracy is another qualifier when attempting to gather quality data. Researchers are constantly asking questions such as, are the results accurate? Is the sample reliable? There is no magic formula but here are a few things to consider when conducting a survey.

**Margin of Error**

There are two measurements that affect data accuracy. The first one is the margin of error (or confidence interval). In sum, this is the positive or negative deviation allowed on the survey results for the sample. In other words, is the difference between the opinion of the respondents and the opinion of an entire population. In order to better understand this statistical explanation, suppose that you set a margin of error of 5% on a study you are conducting about soccer. The results of this survey indicate that 90% of the respondents like to play soccer, a 5% margin of error indicates that you can be sure that between 85% (90%-5%) and 95% (90%+5%) of the entire population likes to play soccer.

The second measurement that affects data accuracy is the confidence level. This measurement indicates how often the percentage of a population actually lies between the boundaries of the margin of error. Following the example above, the confidence level tells you how sure you can be that between 85% and 95% of the population likes to play soccer. Suppose that you choose a 95% confidence level, this interval will indicate that in 95% of the time, between 85% and 95% of the population like to play soccer. A 95% confidence level is standard in quantitative research, since a higher confidence level such as 99% indicates greater accuracy but represents a higher cost.

After understanding the two measurements that affect data accuracy you can use an online calculator to determine the sample size of a population or you could use the formula presented below.

ss= Z 2 * (p) (1-p)

____________

c 2

Where:

Z = Z value (e.g. 1.96 for 95% confidence level)

p = percentage picking a choice, expressed as decimal

(.5 used for sample size needed)

c = confidence interval, expressed as decimal

(e.g., .05 = ±5)

**Screener Questions**

Screener type questions are used to qualify respondents and determine if the respondents are eligible to participate in a research study. For example, when conducting a study on the consumption of alcoholic beverages, a researcher will need respondents who consume alcoholic beverages. Therefore, a screener question should be implemented to eliminate possible respondents that do not consume alcoholic beverages. Consider the question below:

How often do you consume alcoholic beverages?

a) Once a year

b) Once month

c) Every other week

d) Every week

e) Every day

f) I do not consume alcoholic beverages?

If respondents choose option F the survey will be terminated since the respondent will not qualify to participate in the study. This type of question at the beginning of a survey helps increase data quality.

**Data Sanitization**

Data sanitization involves the detection and removal of errors and inconsistencies in a data set due to the incorrect entry of the data. Incorrect or inconsistent data can create a number of problems which can lead to the drawing of false conclusions. Therefore, sanitizing a data set can improve the accuracy of survey results but it has to be done with care in order to avoid problems such as, the loss of important information or valid data.

In sum, there are several elements that dictate survey accuracy. When conducting a survey it is imperative to keep in mind the sample size, margin of error, screening questions and data collection practices to ensure data accuracy and ultimately data quality.