Five Excel workbooks can be downloaded for free which contain worksheets that implement the various tests and analyses described in the rest of this website.

**Real Statistics Examples Part 1A and 1B**: Basic Capabilities, Distributions, T-tests, Chi-square tests, Testing for Normality and Symmetry, Non-parametric tests, Correlation, Reliability**Real Statistics Examples Part 2**: ANOVA, Regression, Survival Analysis, Dealing with Missing Data**Real Statistics Multivariate Examples**: Hotelling T-square tests, MANOVA, Box’s Test, Factor Analysis, Cluster Analysis, Discriminant Analysis, Correspondence Analysis**Real Statistics Time Series Examples**: Time series analysis, Markov Chains, Complex Numbers

All five of these files are compatible with the latest release of the Real Statistics Resource Pack (Release 5.2).

**Download **

Click here on Real Statistics Examples Part 1A to download the Part 1A examples workbook.

Click here on Real Statistics Examples Part 1B to download the Part 1B examples workbook.

Click here on Real Statistics Examples Part 2 to download the Part 2 examples workbook.

Click here on Real Statistics Multivariate Examples to download the multivariate examples workbook.

Click here on Real Statistics Time Series Examples to download the multivariate examples workbook.

Downloading any of these workbooks means that you accept the Real Statistics License Agreement.

**Installation**

Before you use any of these workbooks for the first time, you need to make sure that it synchronized with the **Real Statistics Resource Pack**, the Excel Add-In, as described next.

After you download the Real Statistics Examples Workbook, before you open the workbook for the first time, you should make sure that the Real Statistics Resource Pack has been installed (see Real Statistics Resource Pack Installation). You should close any Excel workbooks that are open and then open the Real Statistics Examples Workbook.

When you open the Real Statistics Examples Workbook for the first time you will likely receive the following messages:

**“This workbook contains links to other data sources”**

If you receive this message, click here to determine how to proceed. If you don’t receive this message then simply use the workbook as you would any Excel spreadsheet.

Alternatively, you may see the following message below the Ribbon and above the Name Box (see Excel User Interface).

**“Security Warning: Automatic update of links has been disabled”**

In this case, click on the **Options** box and then click on **Enable Content**. From this point on you proceed as described above.

**Using the Examples Workbooks**

Once you have download and installed one or both of the examples workbooks, you can click on any tab to access the various examples found in this website. Two tables of contents are provided to help you find the desired example.

The second worksheet in the workbook (labeled **TOC**) is the table of contents for the rest of the workbook. Clicking on any of the entries takes you the specific example described in the corresponding part of this website. The entries in the TOC correspond to the various pages in this website. In this way you can follow along with any of the topics in the website in Excel.

There is a more concise table of contents, which may be found in the first worksheet (labeled **TOC0**). Clicking on any of these entries takes you to the first entry in TOC for that general topic. The entries in TOC0 correspond to the menu items in the Content Menu (that appears on the right sidebar of this website).

Good day Sir.

I would like to thank you for the resources. My question is can Holt-Winters still be effective even if there are only 3 years available of data (given per quarter) to predict the next two years. Also, may I ask what is the best way to solve for missing values given Holt-Winters? I am looking forward for your reply. Thank you.

Hazel,

It really depends on what you mean by “effective”. If you calculate the confidence interval for each of the forecast values, then you will have some idea of how good the forecast is. I have not yet included this for Holt-Winters, although I have included it for ARIMA. I will add this capability in a future release.

If I understand correctly, you have 12 data points (3 years, 4 quarters per year). How much missing data do you have?

Charles

Charles

You have kindly shared an amazing set of resources.

I have enabled Solver and installed Realstats in Excel 2016. On opening any of the examples workbooks, I don’t receive any prompts to update links, but I do see #NAME? wherever formulae rely on supplemental functions and, although I have macros enabled, Ctrl+M doesn’t display the Real Statistics menu.

Any ideas?

BTW, I have completely exited Excel and re-opened an examples workbook, but I’m still seeing #NAME? for {=QSORT(A4:A15)}.

Alan,

This means that the Real Statistics Resource Pack has not been installed on your computer. When you press the key sequence Alt-TI, do you see Solver and RealStats on the list of addins with a check mark next to them? If not, the Real Statistics has not been installed. Also when you enter the formula =VER() you should see the Real Statistics release number.Please reread the installation instructions written on the webpage from which you downloaded the file containing the Real Statistic software.

Charles

Alan,

It sounds like RealStats is not installed. When you press Alt-TI do you see Solver and RealStats on the list of adding with a check mark next to them?

Also what do you see when you enter the formula =VER()

Charles

Charles

I omitted to mention that I had followed your installation instructions carefully, both Solver and RealStats had check marks next to them on the list of addins and, consistent with my earlier posts, the formula =VER() also displayed #NAME?.

I know how difficult it can be to support users with add-ins at the best of times and my regularly updated Windows 10 and Office 2016 seem to be ratcheting up security. Having read http://www.real-statistics.com/free-download/real-statistics-resource-pack/#comment-1079612, I selected Windows Explorer > RealStats.xlam > Properties > General > Unblock. You may need to add this (irreversible per copy of the file) step to your installation instructions.

When I subsequently launched Excel, it kept crashing at the splash screen while opening Solver.xlam, until I tried to “open” RealStats.xlam from Windows Explorer. Excel did not crash and =VER() displayed 5.2 Excel 2013/2016. However, “opening” RealStats.xlam had disabled Solver. When I re-enabled Solver, it started to open and immediately crashed Excel. On re-launching Excel, I received the following error message:

Microsoft Excel

Excel is running into problems with the ‘solver add-in’ add-in. If this keeps happening, disable this add-in and check for available updates. Do you want to disable it now?

Yes No

I selected No and, again, Excel crashed at the splash screen while opening Solver.xlam.

Unfortunately, with up-to-date and fully-repaired Office 2016 and up-to-date Windows 10 Home, RealStats.xlam and Solver.xlam don’t seem to be mutually compatible.

Alan

Alan,

I am not really sure what is causing this problem. I am not seeing it on my computer.

I just had a problem using Real Statistics on my Mac and I followed the following simple steps (translated to the Windows environment) to resolve it:

1. Opened blank Excel worksheet.

2. Disabled RealStats (i.e. pressed Alt-TI and unchecked RealStats, leaving Solver checked)

3. Closed Excel

4. Opened blank Excel worksheet.

5. Enabled Reals (i.e. pressed Alt-TI and checked RealStats)

Everything worked fine after doing these steps, although I don’t know why it worked.

Charles

Thanks for this Excel add-in. I was looking for a software able to make the Wilcoxon test for paired data with Excel and I found this package that has more that I can imagine. I test the package only with the Wilcoxon test and run perfectly.

I usually work with IBM-SPSS but it needs a VPN connection with my University and was impossible for me to do this.

Again thanks a lot for the package.

My suggestion is to do an book, Excel based, with the package and examples explained.

Vicente,

Glad you like the Excel add-in.

I plan to publish a series of books. The first one should come out shortly.

Charles

Hi Charles,

it seems the hyperlinks for Click for Dickey-fuller on Real Statistics Time Series Examples is dis-linked

Mohammed,

Thanks for finding this flaw. I expect to issue a new release of the Real Statistics software within the next week. I will make this correction when I issue the new release. I appreciate your help in improving the Real Statistics service.

Charles

How do i run a GLS (generalized Linear Model) model in your soft ware?

Sam K

Sam,

GLS is incorporated in a number of the Real Statistics data analysis tools. E.g. for ANOVA you can use the GLS model by choosing the Regression option.

What do you want to use GLS for?

Charles

Hi Charles,

Thanks for the great resources. I am being tasked with making a bid analysis conducted by a local municipality for a certain product to be used by that particular local city and with it, are the past data where the participating companies are submitting such as prices they bid, the exchange rates, raw material prices among others. And based on that analysis, I need to make an appropriate prediction of the next bidding outcome, so we can predict competitor’s pricing based on the historical data so we can plan on how should we bid for us to win the contract. Is the Monte Carlo simulation works well with this task or is there any other appropriate approach? Thanks.

Andre,

Glad that you like the Real Statistics resources.

Monte Carlo might be appropriate as well as multiple linear regression. It really depends on the details.

Charles

Hi Charles,

I am working on procurement data of Organization, Where i have complete data. I mean there is no need to draw sample. I need your help to know what kind of statistical functions are useful to analyze this data. Data is like what is purchased from whom at what price, date, with respect to which business unit.

Shailesh,

Do you have specific question for me?

Charles

Hi Charles,

Can you advise how to translate a set of data to a normal distribution using a BLOM transformation, so that Anova or t tests can be used? Do you have a function for this in Excel?

Many thanks

The BLOM transformation takes a value x in your sample S of size n and transforms it to InvNorm((r-c)/(n-2*c+1)) where InvNorm is the inverse standard normal function and r is the rank of x in S. For the Blom transformation c is set to 3/8.

Thus, if say range A1:A20 contains your sample, you can insert the formula Excel formula =RANK.AVG(A1,$A$1:$A$20,1) in cell B1 and the formula =NORM.S.INV((B1-3/8)/(COUNT($A$1:$A$20)+1/4)) in cell C1, and then highlight the range B1:C20 and press Ctrl-D. The range C1:C20 now contains the Blom transformation.

You can also do this in one step, once again assuming that the data is in range A1:A20, by highlighting the range B1:B20 and entering the array formula =NORM.S.INV((RANK.AVG(A1:A20,A1:A20,1)-3/8)/(COUNT(A1:A20)+1/4)). Since this is an array formula, you need to press Ctrl-Shft-Enter. Now range B1:B20 will contain the Blom transformation.

The following article presents some cautions about using this transformation:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2921808/

Another commonly used normal transformation is called Box-Cox. This is supported in the Real Statistics software via the BOXCOX function. See the webpage http://www.real-statistics.com/correlation/box-cox-transformation/box-cox-normal-transformation/

Charles

Hi sir!

Thank you for this wonderful materials.

By the way, do you have excel implementations of X-means. An extended k-mean?

Isaac,

No this capability is not supported at present. k-means is supported.

Charles

Charles,

How would I use an ANOVA Statistical Analysis spreadsheet using eight measurements within a grouping of fifty scenarios? I will be more specific; I am trying to place an ANOVA statistics analysis spreadsheet into my paper using eight different modes of flight (Standing, Taxi, Takeoff, Initial climb, En-route, Maneuvering, Approach and Landing) and when incidences occur during these eight modes of flight during 50 occurrences.

Thanking you in advance,

John

John,

It is hard for me to give a precise answer without more information, but you should be able to use the following webpage to create the ANOVA spreadsheet that you are looking for.

ANOVA

Charles

Here is the data that I need to place into an ANOVA statistical Analysis;

Stages of flight / Number of Fatigue Reported Incidences

A. Standing – 9

B. Taxi – 5

C. Takeoff – 7

D. Initial climb – 4

E. En-route – 7

F. Maneuvering – 2

G. Approach – 13

H. Landing – 3

Total 50

Thanking you in advance,

John

John,

What hypothesis are you trying to test? It doesn’t seem like this is a fit for ANOVA, although it depends on what you are trying to test.

Charles

H1= there is a statistical significance in regards to fatigue related incidents across different phases of flight.

John

Charles, you mentioned that ANOVA may not be the right fit for this type of analysis, would you make a recommendation?

John

John,

It is not clear to me from looking at the data what sort of analysis you want. More specifically, what hypothesis are you trying to test_

Charles

Hi Sir, may i know why this link doesn’t work when i try to download it?

http://www.real-statistics.com/wp-content/uploads/2016/10/Real-Statistics-Examples-Part-1.xlsx

It directs me to do some searching at other pages.

Thanks in advance for response

Best Regards

Hin

Hin,

Sorry about that. I have corrected the problem. Please try the download again.

Charles

sir charles, where can i find a free download of KR20 and item analysis. need your assistance badly.thank u 🙂

Cherryl,

The Real Statistics software contains these capabilities. You can download it for free at

http://www.real-statistics.com/free-download/real-statistics-resource-pack/

Charles

Sir,

Thank you very much for this awesome tool and the very useful explanations. It’s great that you are sharing your knowledge and software with other users.

I’ve downloaded and installed the Real Statistics Resource Pack. However, I can’t find the supplemental function KUDER(R1) in it.

Any suggestions?

Thanks in advance!

Have you typed in the formula that begins with =KUDER(…?

I tried it and it is there. What do you see when you type in the following formula? =VER()

Charles

Hello Charles,

This tool is awesome!! I’m interested in doing some database growth analysis by correlating which table, tablespace, and program on a system that is eating up the database and file system space.

Which tool should I use that can do a linear regression with multiple variables or inputs such as the above listed and something that I can show some correlation and a trend line with fitted plot? Or maybe something that can display graphically all of the details in a chart and good statistical output?

Example

system is total space 800 GIG

DATA

System User Table Tablespace Size(MB) Chg.Size/month

ONE User1 TEST TESTERA 358904.188 28478.466

TWO User2 TEST1 TESTERB 84041.000 6749.671

Jamel,

See my previous response.

Charles

Jamel,

Glad to see that you like the tool.

You can certainly use the tool to do multiple linear regression. You can also model growth (e.g. exponential regression or polynomial regression).

Thank you Charles!

I’ll check the examples and do some more reading on how to use the model growth. My goal is to also do some predictive analysis using statistics with the same multiple variables to not only correlate the data, but also predict what date the storage would get to 100% with those tools too. If you have any more in depth advise, do you mind emailing me please? I’m reading through the stats book to also get the ensure I understand the regression equations too. Thank you!

Real Statistics Time Series Examples workbook the page is missing. Thanks.

Branko,

Sorry about that. You should be able to download the Time Series Examples workbook now.

Charles

Pingback: Real Statistics Release 4.10 | Real Statistics Using Excel

Hi Charles,

Downloaded both example files; example 2 opened fine, however file 1 opens very slowly(1 min) and displays a blank grey sheet when it finishes. Any suggestions?

Using Office 2010 and Windows 7.

Thanks

Dan

My mistake, resolved the problem. Thanks

Hi Charles,

Thanks for the wonderful tools! I am just learning how to use them. Comments and Questions . . .

Comment: For those who have had problems getting your AddIn to work, even though they may have put the file in the suggested location of c:/Users/ . . . /Microsoft/Addins/ , they also need to change the source within Data/Edit Links to point to their own folder rather than your (Charles’) folder. I also had to set Auto-calculate to checked to get it to operate after I made this change.

Question: In Real-Statistics-Examples-Part-2.xlsx, on tab TOC, lines 56-58, you provide 3 Power examples, all referring to a spreadsheet tab that either does not exist or has been re-named in your latest version. Could you tell me which is true, and if the tab exists under a different name, what it’s new name is?

Thank you!

Jerry

Jerry,

Thanks for finding the missing hypoerlink. The link should point to the Anova Power tab (the lines 56-58 should point to cells A1, U1 and Z1, respectively).

I appreciate your help in identifying this problem. I will update the links when I next distribute the workbook.

Charles

Pingback: Real Statistics Release 4.9 | Real Statistics Using Excel

Dear Charles

The website is really useful and instructive.

We conducted an experiment with 2 groups of 8 rats. One was a control group and the other group was treated with a chemical, called “Z” for 4 weeks. We measured a few parameters, including body weights and serum enzyme levels of rats weekly at 5 time points, starting the week after animals were acclimatized and at each subsequent week during the course of treatment with chemical Z.

Our team is divided as to the most appropriate statistical analysis to use to determine if there are significant differences in the body weight between the control and treated rats. One member suggested to use student t-test to compare the 2 groups at each of the 5 time points separately.

Another member suggested to use repeated ANOVA as there is a time course affecting the animals.

We would appreciate your comments. Thank you. Jon.

The Repeated Measures ANOVA seems to be a better fit. Actually you have two factors: a fixed factor Treatment and a repeated measures factor Time. You can learn how to perform these types of analyses on the webpage

http://www.real-statistics.com/anova-repeated-measures/one-between-subjects-factor-and-one-within-subjects-factor/

Note that if you want to consider the serum enzyme levels as well as weight, then you may want to consider using a repeated measures MANOVA.

Charles

Hello Charles,

Thanks you so much for your website, pack and examples.

Just notice that in Real-Statistics-Examples-Part-1 workbook there is no link for the Goal Seek example (cell B30 in TOC tab).

Hope this helps, regards

Christian

Christian,

Thanks for identifying this problem. I will correct this error when I next update that workbook. Yes, this helps. Thanks again.

Charles

Good day Sir,

I was glad to have seen this website, its very helpful.

Thank you!

Hello Dr.,

I found your website very useful. Thanks.

I need step by step formula for ANOVA 3 factors. Is it possible?

Thanks

Abdelkader

See Three Factor Anova

Charles

Thank you so much!!

Sheet “SW Table” n=49 p=0.02 value appears to be incorrect.

RokShox,

Thanks for catching this typo. The value should be .937 instead of .939. I will correct this when I next update the Examples Workbook Part 1, which should happen shortly. I will also change the value given by the SWPROB function. I appreciate your help in impriving the accuracy of the website and software.

Charles

In excel if u have a set of x & y values and next u draw a trend line , by changing trend line type .. exp, log,poly…. r2 value changes .. rite??? Whats the mathematics behind that change , please reply

Yes, if you change from a linear trend line to an exponential trend line, then you are changing from linear regression to exponential regression, i.e. from fitting the data via a straight line versus fitting the data using an exponential curve.

Charles

I’m not able to download the Real Statistics Examples Workbooks. It shows file not found!

Sorry about that. I have now fixed the problem. You should be able to download all three examples workbooks.

Charles

How to calculate x in Beta distribution. If the data table is in cumulative form.

Time C1

1 1

2 1

3 2

4 4

5 4

6 5

7 6

8 8

9 9

10 10

First let’s figure out what the raw data elements are. For your example, these are 1, 3, 4, 4, 6, 7, 8, 8, 9, 10. I did this manually, but you can also do it in Excel as follows:

First convert from cumulative formal to frequency table format. This is done by placing the data in range A2:A11, inserting the formula =A2-A1 in cell B2, highlighting the range B2:B11 and pressing Ctrl-D. You can now convert the frequency data format to raw data format by using the Real Statistic FREQ2RAW function or Frequency Table data analysis tool.

As far as your specific question is concerned you need to provide additional information before I can give a reply. What is x and what is its relationship to the beta distribution?

Charles

Hello i have seen that these downloads are very helpful, i was just confused which example could i use for 3 way ANOVA. My experiment was to test the larvicidal activity of plant extract using 3 solvents, each extract was prepared into 4 concentration, and mortality was observed in 24, 48 and 72 hours interval.

Mortality (%) after

24 hours 48 hours 72 hours

Solvents Concentration

H2O extract 125 ppm

250 ppm

500 ppm

1000 ppm

Ethanol Extract 125 ppm

250 ppm

500 ppm

1000 ppm

Acetate Extract 125 ppm

250 ppm

500 ppm

1000 ppm

This is how the table looks like, i hope you could help me and guide me on how to use the samples and information here. Thank you

Hello Jonathan,

Please see the following webpage for information about how to use the Real Statistics data analysis tool for Three Factor ANOVA.

Real Statistics Three Factor ANOVA

The following webpage may also be helpful

Three Factor ANOVA

Charles

Jonathan,

I see that one of the factors (time) is a repeated measures factor, and so you won’t be able to use the standard three fixed factors model. This situation is not yet described on the website.

Charles

Pingback: Release 4.5 Announcement | Real Statistics Using Excel

I really appreciate you sharing all these wonderful materials.

This website and the resources included are excellent.

They helped me a lot.

Thank you very much.

Salem

A tremendous amount of resources you have available, thanks , has been helpful . A question, you will have the procedure Shapiro – Wilk and modified Mahibbur Govindarajulu 2007

Carlos,

You can run the Shapiro-Wilk test as described on the webpage

Shapiro-Wilk Test

There is the Royston modification as well. See webpage Royston Modification.

Charles

Dr Charles

Could u please provide a link for KR20 using Excel for for 30 students and 20 questions. I just need to calculate its reliability. Thank you and I would appreciate that.

http://www.real-statistics.com/reliability/kuder-richardson-formula-20/

Hi Charles,

Thanks for this powerful resource you’ve provided! I’m wanting to use factor analysis to explore (and refine) an existing MMPI-2 scale to predict vulnerability to developing schizophrenia spectrum disorders at some future point. I’ve installed the resource file, but can’t get the workbooks to work on my Mac using Excel version 14 that came with Office version 11.

Sorry to be one more voice in the chorus of “I can’t make it work”. I’m enthusiastic about using your resource to do something useful with a clinical archival sample of some 750 in-patient cases.

Can you provide additional information about what you mean by “I can’t make it work”. Does this mean that you are you able to use the Real Statistics functions and data analysis tools, but can’t get the examples workbook to work? If so, you probably need to modify the workbook file so that some cell references point to the version of the Real Statistics software on your computer. To do this, look at the following webpage:

http://www.real-statistics.com/free-download/examples-installation/

Charles

Hi Charles,

Thanks for your reply. Sorry, I was entirely vague about what was not working.

You’re right in supposing I had successfully installed the Resource Pack. The problem was with workbook tables which were full of errors that look like ” #Name? “.

Your link to the instructions guiding me through changing the “source” file to RealStats-2011.xlam got the workbooks functioning.

Thanks again!

Thanks for your work.

Gracias por este gran aporte a la Estadística en Excel con los complementos que no los trae la versión original de Excel, ahora podré desarrollar mas aplicaciones de la estadística no paramétrica en particular.

Pedro,

I am very please to see that the Real Statistics software has been useful for you.

Charles

Hi, Charles

As of 9/24, the note above indicates the example spreadsheets are compatible with rel 4.3 of the Resource Pack, although rel 4.4.1 is most current. Typo?

Regards,

Rich

Rich,

Thanks for catching this mistake. I forgot to update that page. It has now been corrected.

Charles

Excellent work. Particularly because of examples, find it very useful in understanding concept. While softwares such as Minitab, SAS, JMP others are very good, having excel based formulae has its own advantages.

I have excel 2013 how can I benefit from these resources?

Brian,

You can down the software for free along with any spreadsheet examples. You can download the examples from the referenced webpage. You can download the software from the Free Download webpage. This software supports Excel 2013.

Charles

Pingback: Rel 4.0 Real Statistics software | Real Statistics Using Excel

Pingback: Rel 3.7 Real Stats Software | Real Statistics Using Excel

Hi Charles, great stuff, thank you.

There is a small fix needed on the Binomial 1 sheet: the x-axis labels on both charts start from 1 and omit 0, but the bar for 1 is the figure for zero. It is most obvious for small values of n where the probability of 0 is high. Everything is shifted over by one.

(To fix Design|Select data|Edit horizontal data labels| =’Binomial 1′!$A$4:$A$24)

I was looking for a way to model successes without resorting to a lookup table. I can take the output of rand() and lookup the cumulative binomial to read off the successes, but it seems a clumsy way of going about it. Short of writing a function I can’t quite see how to do it. The end result will be a Monte Carlo model including annual seasonality, which I might put on my website as a freebie.

Hi Ivor,

Thanks for finding this error in the binomial distribution chart. I have now corrected the referenced webpage to reflect the fact that the first label on the x-axis should be 0 instead of 1. I really appreciate your catching this easy-to-overlook error.

Referring to the second paragraph in your comment, I am not completely sure that I understand what you are trying to achieve, but if you want to generate random values from a binomial distribution, you can do this via a formula like =BINOM.INV(RAND(),p,alpha) as explained on the webpage Simulation.

Charles

Great tools, thanks.

Hi. I am feeling so lucky to find this site and contents. you are a great soul indeed. thanks for all good work. I wish you get all the happiness and wealth in life. once again, thank you very much.

Many Thanks Paul

thanks

Found this site really helpful. Thank you!

Dear Charles,

Great website and great vba program!

Congratulations!

I was just wondering whether you have a file with the description of all the available functions.

Best,

Paulo

Thanks Paul,

There are brief descriptions of all available functions via the

Toolsmenu. More detailed descriptions are given throughout the website.Charles

Hello Charles,

Thank you for sharing all your materials here, these made my teaching job absolutely easy. All the way from the Philippines. More of it, you’re great!

Gem

Thanks. D

I have been looking for something more advanced than Excel for my students. This is it. Thanks for that.

One comment – I have downloaded the Examples Workbook. I notice that in a few cases the formulas refer to a folder which is not where I have saved the addin. For example =’C:\Users\C\AppData\Roaming\Microsoft\AddIns\RealStats.xlam’!VAR2_SIZE(J7)

in worksheet NF3 in cell J24. It is of course easily fixed just by removing the path but I thought you might like to know.

Thanks again, D

Derek,

This is just the path for the Real Statistics Resource Pack on my computer. I give two approaches for eliminating this path name. See the webpage http://www.real-statistics.com/free-download/examples-installation/ for details.

Charles

Thanks a lot Charles., You are the best !

Charles;

THANK YOU!!! I’m running a survey for the second time, 57 items with Likert scale response choices. Your webiste and RealStatistics package mean I can determine if the survey results this time are truly different from earlier results.

Without your work and generousity, I’d be showing histograms and tap dancing! Thank you!

Lisa Allen,

Thanks for your comment. I am delighted that the website and software have been helpful to you.

Charles

Dear Charles,

Great tools, a real improvement over the standard Excel statistics capabilities! However, I think there is a bug in the 2-sample t-test (parametric): I have the impression that you are calculating the test statistic for both cases (equal and unequal variance) using the Welch correction, although it should be used only with unequal variance. If you are using equal sample sizes this doesn’t matter, you get the same t-and t_crit values for both cases (equal and unequal variance). For different sample sizes, however, the t-and t_crit values should be different for equal and unequal variance, which is not the case with your tool. I have verified this using GraphPad and the standard statistcs add-ins for Excel. Would be great to fix the bug!

Kind regards, Chris

Hi Chris,

I am pleased that you like the tools.

You can get more information about the t test algorithm that I am using at the webpage http://www.real-statistics.com/students-t-distribution/two-sample-t-test-uequal-variances/. As you can see from Example 4 on that page I do calculate different values for t-crit for the equal variance and unequal variance tests (even when the sample sizes are the same). The t-stats are the same, however (based on the pooled variance).

As far as I can see, there is no bug. The results seem consistent with other software tools.

If you send me an example where you think that the results are incorrect I will be happy to check.

One thing to be aware of is that Excel’s formula TDIST (or T.DIST, T.DIST.RT or T.DIST.2T) only handles an integer value for the degrees of freedom. Thus Excel treats df = 16.6 as df = 16 (it rounds down to the nearest integer). This shouldn’t impact the test you are referring to since even in this case the two values for the degrees of freedom will be different (unless the variances are exactly equal, in which case the values for the degrees of freedom will be equal, as well as the values for t-crit).

Charles

Chris,

I rechecked my software today, especially after a similar comment from Theo, and you are correct. There is an error in my calculation with unequal samples. I will be correctly the bug and issuing a new bug-fix release shortly.

Charles

hi..i downloaded the realstats 2007 but when i enabled it on my excel it is asking for a password…. i appreciate your positive response on this..thanks

Hi,

You don’t need a password, but you do need to install the software as follows:

1. Select

File > Help|Options > Add-Insand click on theGobutton at the bottom of the window. Alternatively you can simple press Alt-TI (i.e. hold the Alt key down and simultaneously press T followed by I)2. Check the Realstats option on the dialog box that appears and click the

OKbutton.3. If this option doesn’t appear, click on

Browseto find and choose the realstats.xlam file. Then complete step 2 as described above.More info is available at http://www.real-statistics.com/free-download/real-statistics-resource-pack/

Charles

Muchas gracias por compartir su trabajo !! me ha sido de gran ayuda.

Thank you very much for sharing your work! I’ve been a big help.

Renato

Chile

Renato,

I am very pleased that you found the website useful. Thank you for your comment.

Charles

Charles

It appears that the link for the multivariate workbook is not working.

Maybe a blank in the name of the spreadsheet?

As always, thanks for your Excel-supplementing efforts and explanations in yet another release.

Rich

Rich,

I thought I had fixed this (it was an extra blank in the name), but it looks like I didn’t. Thanks for catching it. It should be fixed now.

Charles

I’m really grateful to find your page! It’s terrific and it will help me a lot in my job. Thank you

Hello,

Thank you very much.

The download link on this site contains a superfluous blank.

http://www.real-statistics.com/wp-content/uploads/2013/01/Real%20-Statistics-Multivariate-Examples.xlsx

Removing it makes the download work.

Best regards,

Rainer

Rainer,

Thanks for catching this. I have now removed the blank. I hope this hasn’t caused too much disruption for you and other people.

Charles