Should you do the data analysis by yourself or hire a professional statistician? 


Unless you feel proficient in statistics, I strongly believe that the latter is the right choice. Let me explain why.

Most students who turn to me for help have little if any experience in statistics. Others might find they are unsure of themselves in more challenging assignments. Yet, at their final stage of study are expected to do the statistical analysis on their own. Those who are non-statisticians often find choosing appropriate methods, running the analysis and interpreting the results correctly very difficult. As a result, many dissertation committees have started to allow graduates to hire or consult a professional statistician.

It can be hard to understand why some graduate students spend +100k USD on higher education and are hesitant to spend a few hundred on a good statistician. Not consulting with a professional statistician can be dangerous. It can even jeopardize your graduation.

I mean, would you buy a nice safe car for your family and then put on the cheapest tires?

See my previous blog post where I discuss the price aspects and what to avoid when choosing the right consultant.

Getting value for statistical assistance

When it comes to money we all want the best deal. But as with many things, a balance has to be struck rather than going for the cheapest and sacrificing quality in the process. Statistical services are no different.


A cursory glance on the internet will shoe that what seem to be the exact same services can cost from $10 an hour to $100 or more. You undoubtedly know the answer – it is not the same service! This is true especially for the consulting services. The cost of consulting is driven by two main factors: consultant skill and consultant location. In today’s globalized world it is easy to hire somebody in India to do the work for you for the fraction of the price that you would normally expect in US or other western countries. However, the critical question is: Can he or she provide the same service?

When it comes to statistical consultants, the most important thing is experience and how well the results can be explained to you. A good data analyst is not the same as a good statistical consultant. Let’s admit, statistics is not the easiest field of study. A qualified statistics consultant must be able not only to do the data analysis correctly but also to explain the rationale behind that analysis. Very few statisticians can actually do that!

I have acquired many new clients unhappy with the unsatisfactory work of their previous statisticians. Your investment of time, energy and money is better served by choosing a combination of price and quality. I can offer value, so save your time and energy and work with the best.

So my advice to you:


1. Don’t compare statistical services simply on price. There are important variables to ensure quality. The final output will vary
significantly among the services.

2. When choosing the statistician look for: experience and credentials.

Chi square test

Chi-squared test determines whether there is a significant difference between the expected frequencies and the observed frequencies in different categories. We gather observed frequencies using your data. Then we calculate expected frequencies.

To count expected frequencies, we use following equation:

where n is number of observations, nij is value in row i and column j,  is summary of values in row i, and  is summary of values in column j.

We can do this test in R Studio, which will calculate everything for us without the need to follow steps above). The most important value is p-value. If p-value of chi-square test is bigger than 0,05, it means that difference between answers is not statistically significant.

Assumptions of chi square test are: data should have at least 40 observations and at least 80 % of the observed frequencies should be bigger than 5.

How Does Statistical Consulting Work?

The consultation takes place via email. The first step is to contact me via an email or contact form. To speed up the process, provide me with the data sample and description what you wish to achieve. I will provide you with a quote and turn-around time.

The output you will receive is a clearly written report and interpretation of the results.

You always communicate with me directly. If you have any questions, you can ask me at any time.
I always make sure that clients completely understand the statistics and can defend the results.

To have a better idea how I can help you with statistics, please see my customers’ testimonials here.

logo 2

Contact Info:

Matej Marek, MSc.
STATISTICS HELP s.r.o.

Učňovská 381/10
Prague, Czech republic

(420) 608 515 328
Skype: Call me! - Matej Marek

IČO: 06562370

© 2019 STATISTICS-HELP.COM All Rights Reserved | See also DissertationStatistician.com