Other Statistical Sites

Dr. Gary C. Ramseyer

Dr. Gary C. Ramseyer was an Emeritus Professor of Psychology with a specialty of statistics and measurement at Illinois State University.

Dr. Ramseyer began teaching at Illinois State in 1965 and retired in 1998. Prior to joining ISU, he taught at the University of Iowa in the College of Education and at University High School in Iowa City, Iowa.


Dr. Ramseyer passed away suddenly in the spring of 2012. In his honor, his websites are being managed by his daughter Vicki S. (Ramseyer) Morrow.

Frank Yates Hall

Welcome To The Frank Yates Commemorative Hall!

Frank Yates (1902 - 1994) was a very early computer enthusiast and a 20th century statistics pioneer who contributed to the early development of statistical computing. He became the head of statistics at Rothamsted Experimental Station in 1933 after R.A. Fisher left for University College London. His contributions to the field of statistics were in areas such as the design of experiments, analysis of variance, the balanced incomplete block design and Yate's algorithm. For additional information on Frank Yates, refer to http://www-history.mcs.st-andrews.ac.uk/Biographies/Yates.html

Now displaying the humorous side of Statistics...


Then there was the intriguing story in the early 1940's of the eastern university that had its beginnings in recruiting statistically inclined individuals to serve toward the war effort and offered them free tuition and room and board. The students were housed in elegant Victorian buildings arranged around a courtyard. The instructors hired were the absolute top statisticians from all corners of the USA in the field of multivariate analysis of variance. This project met with such huge success that when the war ended this university was named after their residence halls and the central topic of instruction. What was the name of this university?


My little tale here is obviously fictional but many individuals believe the entire field of statistics is fictional so what else is new? It is interesting that I searched long and hard for another major university in the USA that could be object of a statistical pun like Villanova but came up empty. If any of my readers out there know of a good one, please write me and we will place it in the Gallery.



Have you ever encountered a Statistics Professor...

(a) Who holds an eraser in one hand and chalk in the other gliding from let to right placing material on the blackboard, and then quickly makes it all evaporate before you can take notes on it?

(b) Who proves a theorem with enthused elegance adding an obscure "QED" at the end with a graceful touch, and then NOT have a single student question this strange signature?

(c) Who adopts an expensive 800-page textbook for the course but primarily uses it for the extensive tables and then, to make things worse, only on the days you don't happen to bring it to class?

(d) Who spends HALF of the semester on the first third of the content for the course and the LAST WEEK of the semester on the last third of the content?

(e) Who regularly refers to "calculus" as the explanation of a statistical principle but knows full well that 90% of the class has never taken the course let alone heard of the word?

Yes these behaviors should ring a bell for those who have taken a course in the beloved discipline of statistics. I would be remiss if I did not admit to committing several of these quirky actions myself as an instructor. Bet you can enlarge this list by many more if you think hard enough.


A beautiful young woman invited a brilliant statistician friend to her Company dinner-dance. The invitation stated that she could either bring her spouse or her significant other as a guest. having just met this chap and being unmarried, she felt certain that he would more than fill the bill since all statisticians are by definition statistically significant. When they arrived at the door, the maitre d' inquired as to the status of her escort. She smiled and promptly introduced him as an up-and-coming statistician that was her significant other for the evening. The maitre d' was stunned and his face grew red. He finally stammered in an embarrassed tone of voice, "I am so sorry madam, we cannot admit your friend.


You have to empathize with this young lady on the shocking news from the maitre d'. Her whole evening was no doubt ruined. But you know something, the maitre d' was absolutely correct. Statistical significance does not imply practical significance but the converse is true, practical significance does indeed imply statistical significance. This fact is probably not emphasized enough in an introductory statistics class. Let me give you a simple example that illustrates this principle. A researcher was testing a two-tailed hypothesis that all the 6th graders in the city's school system was from a population that had a mean IQ of 100 (i.e., H0: µ = 100 vs. H1: µ < 100 and H2µ > 100). The sample data were as follows: N = 1600, Bar X = 101, s = 20. Now computing the est. standard error of the mean we obtain s/sq.rt.(N - 1) = 20/sq.rt(1600 - 1) = 20/40 = .5. Finally computing the z-statistics we have z = (101 - 100)/.5 = 2.00. Holy Toledo! This observed z is significant at the .05 level and we can reject H0 and accept. H2: µ > 100. Before the researcher gets all excited though and give high-fives to all the principals of the schools, he should reflect on the whole situation. He had a HUGE sample which led to a dinky standard error and this in turn produced a significant z for a difference of ONE between the hypothesized and observed IQ means. You must be kidding---No we are not! Most sane people would consider a difference of one in means as trivial and therefore practically unimportant in any meaningful endeavor that would be performed on this population. Thus the moral of the story is that when a researcher is blessed with a very large sample(s), and a hypothesis test results in a statistically significant result, ALWAYS reflect on the dependent variable(s) to make sure that the mean difference or whatever produces a judged practical value in real life. I realize this is somewhat subjective but it is a vital step in the process of interpreting a hypothesis test. Now all the young woman has to do the next time she invites this talented statistician out is to tell the maitre d' she did NOT use a very large sample!!!



There once was a statistician named Maximilian.
Who gained his fame as a nonparametrician
He fondly professed his preference for rank
Though most of his colleagues thought that this stank
And after many failed rejections of the null he was tagged an antiquarian.

For those of you that don't know, nonparametric tests were the dance craze of the decades of the 50's and 60's. Everyone wanted to do them because they mainly involved simple ranks and were easy to compute. These boosters presumed the assumptions of parametric tests (Like the t- and the F-test) were never really satisfied and hence the tests not valid. Then the advent of the high-speed computer in the 80's and 80's and Monte Carlo studies were performed on a variety of violations of these assumptions. The effect on Power and the probability of a Type I error was then studied. And lo and behold, in most studies, moderate violations of the assumptions did not affect these two criteria much (with n's of the treatment groups assumed equal or nearly equal). in other words, many parametric tests were robust. Well, the people tooted their Vuvuzelas and jumped on the bandwagon of parametric tests in the 90's. Also, an added bonus was an increase in the power of the test and the finding of more significant results. Wow! Researchers did love this! To commemorate this event, I wrote the most recent of my limericks.



Why did four groups of chickens cross the road?

Because ANOVA's were done on the other side.

Why were ANOVA's done on the other side?

Because only t-tests were done on the first side!!!

Give the chickens credit for a high degree of statistical acumen here. When you have 3 or more treatment groups and you are N>OT interested in apriori comparisons, you should do an overall ANOVA first to see whether you have any differences at all among the treatment population means. Please note the presence of FOUR Groups of chickens in this study!



In this year of massive recalls, it was just heard on the evening news that there was a recall of certain hip replacements. What? You've got to be kidding me as John McEnroe would shout!!

But REJOICE all you statisticians, this finally gives us the right to RECALL many published significant studies from journals that were later proven to be nonsignificant!!!

This has always been one of my pet peeves...the bias in any discipline toward publishing ONLY statistically significant results in its journals. It is rare indeed to find a printed article where the results are all nonsignificant. Just think of the revolutionary effect and balance in research findings these recalls wold produce more specifically, journals would be required to print in subsequent issues, updates and or outright statements nullifying some former conclusions. Wow, how this would shake up the research enterprise in many fields!!!



Where do statisticians have accidents with their cars more often?

Research has shown that it is the MEDIAN of multi-lane interstate highways where the grass is MODE short. What does this finding MEAN? It suggests that statisticians are middle-of-the-road people who detest being an outlier in the ditch.

Well, it appears that statisticians are AVERAGE people after all. Thanks to David J. Bell of Western Michigan University for giving me the idea for this little funnyism.



Perhaps everyone should lighten up on the President.

The Stimulus is working!

Why, just two months ago he got jobs for 63 Republicans!!

Moreover, he praised the original t-Party (William Sealy Gossett's) for declaring this job increment statistically significant at the .001 level!!

It also should be mentioned that these Republicans had abandoned looking for work so this is a DOUBLE achievement for the President. Thanks go out to my former student, Dennis Wright, the Pin-Man of Iowa Homecoming Football, for helping me construct this joke with some subtle changes.



Let us ponder the unthinkable. Suppose the Normal Curve as we know it had never been discovered by Gauss? What would the statistical world of today look like?

(1) Of grave concern, sampling distributions of statistics would never APPROACH a normal distribution. What a revolting and depressing state of affairs

(2) The Bivariate Normal Curve in 3D (i.e. the true bell curve) would never appear and be rotated on Internet web pages. Additionally, 3D-movies of this action would never be made and the 3D glasses industry would suffer immeasurably.

(3) The ubiquitous one-page standard normal curve table of selected percentiles would not appear in any statistics textbook. This table has been the hallmark of all statistics texts and statisticians would go start raving mad not seeing this gem.

(4) The Statistical tee shirt business would be defunct.

(5) Percentiles of sampling distributions of every imaginable statistic would all have to be developed individually and tabled because of (1). This would lead to monstrously thick textbooks with half of the book devoted just to tables. Oh my, the cost of a book to a student would be $500 or more! Also, how would a student lug this to class?

(6) AS the df-value of a t-curve increased, the tails of the curve would shrivel up and go to nothing and the points of inflection would disappear. That is, the curve would have domed top and vertical sides. The end result would be nothing that you would recognize (maybe look like a garbage can) and certainly far removed from a normal curve. But wait a minute, we don't have a normal curve anyway so who cares what we end up with?

(7) All curves or distributions would now be labelled ABNORMAL for a start and a whole new nomenclature for these would have to be developed to lend descriptive qualities and specificity to their appearance. What a nightmare and polyglot of new names!

(8) The nasty-looking formula for the probability density function of the normal curve along with its contained venerable "e" (the base of the natural logarithm system) would no longer appear in textbooks and scholarly works. This would be a horrific loss because an instructor could no longer flash this formula on the board for shock effect when introducing the normal curve to show students this curve is really not very "normal" mathematically.

(9) And last, but not least, the multivariate normal distribution and all its applications would drop from the statistical scene. This could be the most heartbreaking loss of all since matrix algebra is the heart and sole of this technique with the necessity of the variance-covariance matrix employed in lieu of just a single variance in (8). In my estimation, this is the most elegant application of matrix algebra in all of mathematics and enables the field of multivariate analysis to be explored with concise clarity and understanding.

Well folks the above is just a start and contains only nine items because I did not want to mimic "David Letterman's Top Ten." You could add many more to this list. Putting it bluntly, the absence of the Normal Curve would be STATISTOCA: ARMAGEDDON. So, hug every Normal Curve you encounter and thank every statistician that has perpetuated its presence. The Normal Curve and Statistical Techniques are forever inseparable!


The Normal Curve in its critique
Is beautifully symmetrical and sleek.
Sometimes it is skinny and tall
Other times fat and real small.
But with it the data will always speak.

Yes, the Normal Curve, is the heart and soul of statistical methods. I get goose bumps just thinking about the number of times I taught the ins and outs of probabilities under this elegant graphic. I want to thank Jared of Tennessee Tech University for sending the original working of this limerick which I modified slightly.



Where do married couples who are both statisticians sleep?
On a nonsingular variance-covariance matrix!

And why is this so comfortable?
Because the inner springs have the variances and covariances symmetrically arranged!

Oh, the sheer beauty and simplicity of stating the probability density function of a multivariate normal distribution using the variance-covariance matrix in the formula is so neat. Also, the the thought of using matrices to conceptualize p-variate multivariate tests when p>2 without the use of pictures is one of the most impressive revelations I have encountered in mathematics.



THE MURPHEY EXPERIMENT: One day a professor named Murphy wanted to demonstrate the laws of probability to his statistics class. He had 30 of his students spread peanut butter on slices of bread, then toss the bread in a flipping fashion into the air to see if roughly half (the expected value = 30/2 = 15) would fall dry side up and half would fall on the buttered side, assuming that this sample would be representative of a sample distribution of bread tosses. As it turned out, 29 slices landed peanut butter side down on the floor and the 30th slice stuck to the ceiling. GASP!

MURPHY'S REPLICATED EXPERIMENT: A true scientist, Murphy conducted the experiment a second time after extensive cleaning of the room. All of the bread landed buttered side up! Murphy consulted his old statistics professor for counsel on what he was sure was a very rare event. His old professor didn't feel qualified to deal with the question, and conferred with colleagues of the International Society of Statistics and Probability. After months of waiting, the scholarly verdict was relayed to Murphy..."You buttered the bread on the wrong side!!"

What else could go wrong for professor Murphy? Thanks to Doc Kenn Finsteun once again from San Antonio, Texas for the funniest and most extreme illustration of Murphy's Law that I have seen.



"Most people give you an anticipatory grin when you mention a STATISTIC, frown doubtingly when you mention the plural STATISTICS, and grunt and groan in a gurgle when you mention A STATISTICS COURSE." -- Dr. Gary C. Ramseyer

OK, what could I possibly mean by these progressive behaviors? The singular use of "a statistic" usually elicits from a person a happy thought that he or she is going to hear a very intriguing fact such as only 0.4% of all the at bats are grand slam homers in Major League Baseball; the plural of "statistics" firms the chin angularly and reminds one of conflicting conclusions in medical treatments that change every six months; and finally "a statistics course" drives a person hysterically up a wall when TOO MANY statistical methods are taught but NOT ENOUGH situational buckets of scores are presented to analyze.


Did you hear about the big robbery several weeks ago at the Museum of Science and Industry in downtown Chicago?

Two masked and armed men stole six ORDINANTS near the middle of the large replica of the NORMAL distribution and the whole thing collapsed like a circus tent. When the maintenance staff discovered the disaster they were scared half silly. They finally decided to take all the remaining ORDINANTS, cut them into equal lengths, and reassemble everything into a model RECTANGULAR distribution. The next week NOT A SOLE visited the exhibit because it was UNIFORM and boring!

This calamity certainly has saddened the entire statistical world. No longer will visitors witness the two points of inflexion but will also miss the phenomenon of both tail-line curves approaching the score scale but never quite touching it. I think a donation would certainly be in order here for the restoration of the original exhibit to bring back the feeling of awe and elegance associated with the Normal Curve!