Statistical Analysis with Excel For Dummies, 4th Edition

I am a big fan of Joseph Schmuller Statistical Analysis for Dummies books. Earlier I had reviewed his first edition that catered to Excel 2003. His new 4th edition covering Excel 2016 is just as good. It is also more expansive. The first edition had 392 pages. The current 4th edition comes in at 527 pages. The author has added two new chapters, including a very interesting one on non-parametric statistics. And, he treats many of the subjects in greater detail. He also shows some of the additional formulas and capabilities of Excel 2016 that were not included in Excel 2003. Below is the text of my original review of his first edition that is just as relevant today.

I bought this book because I am currently researching multiplicity or how to test multiple hypotheses. This is one of the trickiest and most interesting areas of statistics. I am an advanced Excel user and a reasonably proficient autodidact statistician. Additionally, I have stuck with Excel 2003 as that is the Excel format I have used for decades and saw no value in relearning Excel from nearly scratch to adapt to the newer versions.

In view of the above, I was delighted in finding this early version of this book (copyright 2005) that demonstrates all the hands on statistical examples in the Excel 2003 version (the version I use!).

As a result, I have learned a ton about multiplicity and how to deal with it. The Excel Add-in has terrible semantics regarding all the ANOVA techniques. For instance to do a Two-Way Between Group ANOVA, you have to use in Excel the ANOVA: Two-Factor with Replication (which actually describes a different method). And, Schmuller does an excellent job of guiding you through Excel's spurious ANOVA semantics.

The book covers a surprising amount of statistics. Once you complete reading, studying, and practicing the examples within this book, you will know a great deal about statistics. The latter should provide an adequate foundation for any upper level degree in social sciences.

Schmuller does an excellent job of outlining how much you can do in statistical analysis without earmarked very expensive software such as Matlab, SAS, SPSS or really cryptic ones such as R. In the end, you will be surprised with Excel's capabilities. Schmuller shows some pretty advanced quantitative methods including: the mentioned ANOVA techniques, probability, and statistical distributions sections. He even shows you how to conduct Monte Carlo simulation with regular Excel! That can save you a ton of money not having to buy another set of Monte Carlo simulation earmarked software such as Crystal Ball or @Risk which would set you back over $1,000.

If you must use Excel for analyses this book is a great place to start. It will show you how to make the best of the statistical tools that ship with Excel. In addition to showing you how to access and use the tools this book gives a great overview of how to work with formulas and how to do basic math. The coverage of the math and theory that goes with the statistics is approachable and often very good.

While all this is good, as a professional statistician I found myself thinking but what about checking the assumptions for the tests. Modern free tools like SAS University Edition let you point and click your way through these statistics and many more and they automatically kick out diagnostic graphics and checks on the assumptions that must be true to trust your results. Excel does a lousy job of helping novices check assumptions and because of that fact this book suffers.

So, if you must use Excel pick up this book. Just be sure to talk with an expert and have them help you check to make sure that the inferences that Excel is telling you are actually valid.

I have the 3rd edition ([[ASIN:1118464311 Statistical Analysis with Excel For Dummies]]). The 3rd edition runs 480 pages, this new 4th edition runs 20 pages more (500 pages). Part 1 ("Getting Started...") runs from page 7 to 62 in the 3rd edition and 7 to 60 in this 4th edition. Part 2 ("Describing Data") runs from page 63 to 184 in the 3rd edition and 61 to 182 in the 4th edition. Part 3 ("Drawing Conclusions from Data") runs from page 185 to 352 in the 3rd edition and 375 in the 4th edition. Part 4 ("Probability") runs from page 353 to 412 in the 3rd edition and 377 to 436 in the 4th edition. And Part 3 ("The Part Of Tens") runs from page 413 to 450 in the 3rd edition and 437 to 470 in the 4th edition.

This means Part 1 is pretty much the same in both editions (56 versus 54 pages); Part 2 is 122 pages in both editions; Part 3 is 168 pages in the 3rd edition. but 194 in the 4th (meaning Part 3 is where the book has been most improved upon; there are two new chapters (read more, below)); Part 4 is 60 pages in both editions; and Part 5 is four pages shorter in the 4th edition (34 versus 38). The two appendices are the same in content and length. Both editions show two bonus appendices ("When Your Data Live Elsewhere" and "Tips For Teachers (And Learners)"), but they're not in the book (I'm guessing you have to go online).

There is a difference and perhaps a significant one (not counting the software interface differences between MS Excel 2016 versus MS Excel 2013 (which I don't think are all that significant)): the addition of two new chapters - Chapter 16 ("It's About Time") and Chapter 17 ("Non-Parametric Statistics"). Chapter 16 was interesting and I think a case can be made that its inclusion makes it worthwhile to upgrade (frankly, though, Chapter 17 is beyond my ken at this point so I can't say whether it makes an upgrade worthwhile). So here's where I come down on this new 4th edition: if you have the 3rd edition, are comfortable with Excel 2013, have comfortably made (or think you can make) the transition to Excel 2016, and can do without the two new chapters, then I'm not sure the book is worth the upgrade. But in any other case I highly recommend you not try to save a few bucks by buying the 3rd edition and purchase this book ($27.60 versus $21.60 (Amazon prices at the time of this review)).

Bottom line, the majority of stats most of us do do not require specialized programs such as Mini-Tab, SPSS, etc. They are no doubt fantastic tools, but they are pricey. So if you can get away with doing most of your stats work with Excel (a tool that most of us have even if we've never really used it much) that's a real bargain. Excel is also a great way to go because many of us learned stats on a TI-84 or Casio handheld and one of the things we learned was that working with a hand-held has severe limitations.

As such, this book (in any of its editions) is a fantastic resource (I say this as someone who counts himself severely math-challenged). Author Schmuller does an excellent job of walking the reader through the concept of the stat and how the stat is worked with/through in Excel. The author jokingly states he covers every stat function Excel offers but one ("Fourier Analysis") which to my mind means Excel can handle pretty much anything a non-specialist can throw at it (and some) and that the author helps you through each and every one (minus one). That, to me, is value for the money.

If my dream to tackle a doctorate-level degree comes to fruition, stats will once again present me with a tough challenge. I'm glad this book will be on my desk helping me along the way.

5 Statistically-Significant Stars! Nice job.

This author really understands how to convey unfamiliar information. He walks you through the appropriate conceptualization, never missing a step. He explicitly sets aside more complex information, but gets back to it when you are ready for it. Besides he is very very funny. A real treat. I used to teach statistics, and I envy his crystal clear presentations.

The author does a good job of explaining theory. Not such a good job on implementation. Sometimes it seemed as if he felt it was time to end a section and kind of dropped off. I ended up going online for clarification elsewhere. The Excel examples are generally good. The Excel illustrations are hard to read-too small- and I had trouble copying numbers for examples even with my glasses even though I generally do not need them when I read and work on my computer. I have used other Excel books that offered the example spreadsheets online.

Not enough step by step examples. Too many words. Especially when there are so many functions to cover. To me, words dont help, I need examples.

Nicely written to help acclimate me to the world of statistics

At close to 600 pages and written in an easy to follow manner and geared now to Excel 2016. While the reader should have at least an elementary knowledge of statistics, the examples are geared to the beginner with just basic knowledge of Excel and standard elementary statistics.

The biggest addition I can make to the other fine reviews is that this book and the use of Excel for Data Munging provides a good resource for those venturing out into the modern obsession of machine learning. Excel with the included tools provides a good resource for examining and manipulating a data source before venturing into more powerful Statistical and Machine Learning tools like R and Python.

I love how the book clarifies how to use Excel's statistical analysis tools and statistics functions. I am getting ready to return to college and this book will be an awesome adjunct to my assigned text for Statistics for Psychology.

Timely arrival. Use for reference.