# Web Analytics are our data accurate

[introduction] in the long history of

website analysis, asked the most is the problem of "your data accurate?" the website analysis data is accurate or not, how to treat the deviation data analysis of potential sites, this paper will give the answer.

[text]

is accurate and accurate in Chinese. It’s a synonym. We can mix it in spoken English. So is English. Accurate and precise are also casual and blurt out. However, since there are two words in existence, but not in the process of evolution of the text of one of the dying, it shows that they are still subtle differences. In fact, accurate and precise is definitely not the same concept, they are in engineering, statistics and many other sciences are distinguished strictly for the website analyzed this emerging discipline, so completely.

, let’s look at the difference between accuracy and precision, and then see if the site analysis tools can be accurate or accurate, or that two are prepared.

what is accurate, what is accurate

?Excellent on accurate and precise interpretation of Wikipedia

, a classic. Here I use Chinese to pay tribute to it: refers to the phenomenon of accurate measurements or relative facts between the discrete degree is small, our spoken English "close to the truth and facts"; precisely refers to under the same conditions, the phenomenon or the measurement value can repeatedly simulate low dispersion degree, also is our oral English "and so forth," etc.. The two map from the Wikipedia classic, reference:

Figure 1: This refers to relatively high accuracy, but with relatively low accuracy,

Figure 2: This refers to a relatively high degree of accuracy, but with relatively low accuracy

The red circle in the two chart above

represents the facts. You can see in Figure 1, measured values around the center, although the distribution of discrete, but it can be seen that the average distribution of them must be in the heart circle (or that average of multiple measurements is consistent with the fact that), so can be called accurate, but because of the discrete and not as accurate. In Figure 2, the measurements are clearly off the center of the circle (the average of the measurements can not be in the center of the circle), so it cannot be called accurate, but it can be called precise, because the reproducibility of the measurements is very low. This is an excellent explanation for accuracy and precision.

if we establish accurate and precise matrices in two different dimensions, we can get the following diagram:

, figure 3: accurate and accurate matrices (source: www.wellesley.edu)

left up >