Know-How: Don´t be afraid of statistics!

If you start a survey not only the survey is important but also the result. To process the results, statistic values are of great importance. Even if statistics seem to be complicated, they are, in fact, simple. Statistics turn huge amounts of numbers into a manageable amount and define probabilities for "events" that have not been analyzed.

Among others statistical values, values like count, frequency, median, mode, standard deviation, absolute deviation, range, variance, kurtosis, poisson and skew are of great interest. This values are explained below:

Count: Here the number of people who answered is displayed.

Frequency: Here the percentage of people who answered is displayed.

Average: The average is the sum of values, divided by the number of values.

For example, if you ask seven people, how many people are employed in their company, and you get the answers 10, 25, 30, 35, 40 and 50, then the average is 31,4 (10 + 25 + 30 + 30+ 35 + 40 + 50 = 220 / 7 = 31,4).

Median: The median is the value in the center after all the values have been sorted in ascending order. In the above example the median is 30. If you have an even number of values, you can choose one of the values in the center. If this values are not the same you take the average of these values.

Mode: The mode is the most represented value. In our example the mode is 30 ( it just happens to be the same value as the median).

Standard deviation: The standard deviation curve shows how the scores are spread out. A high standard deviation indicates that values deviate more from the mean - are more spread out - than samples with a smaller standard deviation.

Absolute deviation: Here the absolute values of the differences (and not the differences) from the mean are summed, because adding the positve and negative values of differences in a distribution from its mean, will always be 0. The absolute value of negative 5 (-5) is 5, just as the absolute value of positive 5 is 5.

Range: The range is the difference between the highest and lowest data point. For example, looking at the answer values 10, 25, 30, 35, 40, 50, the range is 50 - 10, which is 40.

Variance: The variance is the spread of the values from the mean. The square root of the variance is the standard deviation (see above).

Kurtosis: The kurtosis is a measure of the peakedness or flatness of a distribution. The kurtosis for normal distribution is 0. The more the values deviate positive or negative from 0, the more definitive is the measured skew. The kurtosis helps you to find out the "right" distribution.

Poisson: The poisson distribution is used to find the probability of the number of events that occurr in a specific time period.

Skew: The skew relates to the asymmetry of a distribution around its average. Positive skew means that the distribution has a long tail in the positive direction (to the right). Negative skew means that the distribution has a long tail in the negative direction (to the left).

Using these values mass appearances can be analyzed and presented in numbers. By means of numbers, it is easier to compare results. A numerical value is clearer than thousands of values that may come out when launching a survey. It is also obvious that the average is more significant than a number of numerical values. For example, if you ask 30 people for their age, you get a better idea if you know the average value, than having 30 numbers in front of you.

So don´t be afraid of these complicated statistical values, because they help you to analyze and evaluate your survey results.