# Measure the position

Don't hesitate to turn to our statisticians. We are proud to offer you quality statistical services supported by 100% money back guarantee. Regardless of the test and software you choose, you can expect from us:
• comprehensive task evaluation and quick quote
• custom approach to your statistical project
• accurate data analysis and detailed interpretation
• constant project progress updates
• free adjustments and timely delivery
Let us take care of your data!

It is said that from a set of observations, we always try to understand the specific position of a value comparing to other values. My question is related to percentiles. Are percentiles not enough to measure the position?

### 1 comment to Measure the position

• Jack

Your question is quite interesting. To answer of your question, first I need to know about your objectives as well as nature of data sets. It is true that in basic SPSS research, percentiles play an important role; however, sometimes you have to use other measures to know the position of your value. I simply use percentiles, quartiles and standard scores (z-score).

When do we use percentiles? If we find that all the values of a data set are sorted from the smallest to the largest or vice versa, we use percentiles. All the values are divided in a rank-ordered set of elements by 100 equal parts and we try to assume in which part we can find the most significant result.

On the other hand, in quartile, we divided the whole data sets into four major parts equally which is called first, second, third and fourth quartiles and those can be identified by Q1, Q2, Q3 and Q4 or something like that. The difference between percentiles and quartiles is very clear that when you try to understand very scattered data (where data range is high, means minimum and maximum data range is high), quartile can help us for better interpretation. However, another measure of position is standard score or z-score or aka. If you subtract the mean of population from the value of a element, you will find a score which should be divided by standard deviation is called z-score. From this z-score, you can easily identify the exact position of your desired value in your SPSS research. For better clarification, you can use in-build SPSS help option.

I should say that sometimes percentile is enough in your SPSS research. But everything depends of your study objectives. First look at your objective and try to understand what you want from this. You will find the answer from your own.