My father once bought us the entire set of encyclopedia britannica. The books were suppose to serve as our information source. Our world wide web had not been invented as of yet. (Am I dating myself right now? Probably so lol) In any case, if we had a question about our school work, the world we live in, or anything at all- we had to look it up. Over time, I began to resent my father for not answer our questions. Why couldn’t he just answer the question? In my mind my father knew everything. And he did to some extent because he read BOOKS. He explored the world he lived and found answers to his questions. So I really took it to heart that he withheld information or so it seemed at the moment. And let me tell you, the answer was never in those encylopedias. I would be stuck on book M for an hour looking for the answer. Where is the freaken answer? The word I am looking for is not even in the book!

I now understand the message my father was trying to teach me. The answer is a living being that grows over time. Different contributors give you a perspective of what the answer could be- but you never get the full picture. You need  as many BOOKS and resources to get a better sense of a topic. Now how does this relate to statistics? Your statistics teacher is a limited source of information. I will tell you right now, I do not teach a lecture to encompass every aspect of a topic. I choose a particular angle and teach from a basic level of understanding. So you have to decide your limits as far as fact finding. Some people focus on artificial intelligence. Others stop at inferential. You never go into a statistics class or any where and assume the information  will be handed to you in a fast food bag for instant consumption. This is not McDonalds , Burger King, or White Castles. Read your books but also research the truth beyond your immediate resources.

-Moore to follow Amy

Hello people! I have not written a blog in a good minute.  I was inspired this morning to write about the steps leading up to analysis. So I have consulted quite a few students up to this point. And I made the observation that some research questions are formulated without considering how to collect the data. I am not speaking what the data will look like when you finally analyze it. Rather we should focus on random versus non-random sampling. And the key issue is this thing called bias. When you try to show statistical significance, you want the results to represent what is actually happening in the population. Introduce bias in your data, distort the claim in your paper. Random sampling ensures that the data is collected without bias. There are certain statistical tests that you can not run if your data was not collected at random.

I keep throwing around this work random. What I am I talking about here? You want your study participants to have an equal likelihood of being selected for the study. Probability sampling ensures each person in your sample had an equal likelihood of being chosen for the study. Now if you have a convenience sample, data collected at a particular place and time- you can not generalize about the larger true population. Convenience sampling is not based on probability and you can not analyze the data using parametric tests.

I suggest getting familiar with the way you collect the data before the situation is irreversible.

-Moore to follow Amy

Dear All,

As an undergraduate student at Rutgers, I had no idea how to acquire marketable skills for the outside world. Technically, college is part of the outside world but it’s really not OUT-side. It’s more of a parallel universe. You are not in the workforce YET.  That is why I have dedicated some of my time to helping students gain skills they can put on their resume. Now I am strictly referring to statistics topics. If you want to be an analyst, you need to learn popular statistical software languages. If you want to brush -up on statistics, I can refresh your memory on the language. You should never go into an interview ‘cold’ without a few tools in your arsenal.

-Moore to follow Amy

How many times have you sought help for a statistics assignment or dissertation? How many times have you heard that the price for tutoring is out of this world? Well, I am offering to help people for free if they buy a tee shirt by November 3rd. So the tee shirt is 20 bucks. So for that amount of money you get a 20 dollar shirt AND a free consult by yours truly.

Chi-square tests for categorical data will reveal if there is an association between the variables in your contingency table. However, the test statistic will not tell you where dependencies exist within the cells of your table. People typically illustrate count data with bar charts, pie charts, and histograms. And those illustrations are great for a beginning-level course. The natural inclination is to show the percentages and quantities for each category. But I want you to imagine an graph that will tell you which levels show the greatest associations between the two variables- but with COLOR!  

Mosiac graphs can be generated with R Code. Here is an example:

The associations are represented by the different colored boxes. If there is no association for that particular variable level the box stays grey. Pretty neat isn’t it?! If you would like to learn more please contact me for an appointment.

-Moore to Follow Amy

The phrase Kendall Tau is not typically introduced in your basic statistics class. But if you were performing a correlation for your  dissertation defense- make sure you don’t need one. As a matter a fact, do me a favor…If you are working with a correlation right now…Classify the variables in your correlation as one of the following : interval, ratio, nominal, or ordinal. If your variables are ordinal a Kendall Tau is more appropriate than a Pearson.   The Kendall Tau is a non-parametric correlation for ordinal variables. Ranked Correlations for ranked data!

The Pearson’s correlation coefficient is more appropriate for continuous variables (interval or ratio). When you are studying linear correlation, your variables should be aligned with a normal distribution or at the very least the linear family. And this is my perspective, but count data is not part of the linear family.

There are other non-parametric correlation tests. Know your variable type.

-Moore to Follow Amy

My company tee shirt is in the shop on my site. It contains a butterfly symbol which I personally chose because it is a reminder of my late father Gary Moore. My Dad was not a mathematician, instead he practice law. What is the connection to butterflies? One day I was talking to my Dad while in school for my PHD in Biostatistics. And I was telling him about my obstacles and stumbling blocks- and a couple of beautiful butterflies came in my path. I told my Dad and he said ‘yes I may have sent them to you’.  Now it was not butterfly season, it was the middle of winter and it caught me off guard. But at that moment, the butterflies symbolized hope and perseverance. And my Dad was the epitome of hope, strength, and determination.

In memory of my late father, Gary Moore, I would like to offer ‘hope’ to people who are studying statistics  or fulfilling their statistics requirements. I am offering to tutor people for 1 hour ($60 dollar value)  if you purchase one of my tee shirts ($20 to $25 value) in the shop. I hope you enjoy it!

-Moore to Follow Amy