There is actually nothing really impossible in Excel, but many things are really really hard to do with Excel in an automated way (adding fake data serias to show the connecting lines, and more fake data series to show the p-values; calculation of the correct positions is cumebrsome and the correct formatting is hell). I would add the extra annotations simply as graphical objects (lines and text), either in Excel itself with a graphics software in the exported graphic file. (Many people here do it in Powerpoint, but that's not my favorite because it gets messy when you have to prepare a figure for publication in a given file format (say, tiff), size, and resolution... this is better done in appropriate software like Photoshop and alikes).

Regarding the "normalized gene expressions", and without knowing much about your specific experiment and data: I hope you are aware that such measures ([relative/normalized] expression) are usually not normal distributed, and that therefore "mean", "SEM" and "t-test" are usually not sensible. Often, it is reasonable to assume a log-normal distribution here, so you should do all your statistics on the log-expression values (where you can sensibly calculate means and SEMs and t-tests).

Also note that a bar plot is the (unfortunately most common but) worst possible way to present expression data. If your y-axis is "expression", note that the uncertainty is not symmetric around the location measure, so showing only one-sided error bars would be wrong (or at best incomplete). In any case, make sure that the bars extend to a meaningfur reference value (this may or may not be 1 for "expression" and 0 for "log expression", depends on the normalization and meaning of the data). A far better alternative is to show dots instead of bars (either a single dot/symbol at the position of the location measure or showing the complete data with each dot representing an individual measurement. You can then still use error bars to indicate SEMs or (better!) confidence intervals. Wheras bar plot highlight the absolute (mean) values compared to the reference (where often simply 0 is used, as in the attached example), the dot plots highlight the differences between the groups, what is typically the main research question. If you have >20 or so values per group you could also go for a box(-and-whisker-)plot (unfortunately not really available in Excel; workarounds and tricks known to me fail when values are negative). I attached an example showing the same data as "1d scatterplot" (individual data), "dot plot" (with error bars) and "box plot".

There is actually nothing really impossible in Excel, but many things are really really hard to do with Excel in an automated way (adding fake data serias to show the connecting lines, and more fake data series to show the p-values; calculation of the correct positions is cumebrsome and the correct formatting is hell). I would add the extra annotations simply as graphical objects (lines and text), either in Excel itself with a graphics software in the exported graphic file. (Many people here do it in Powerpoint, but that's not my favorite because it gets messy when you have to prepare a figure for publication in a given file format (say, tiff), size, and resolution... this is better done in appropriate software like Photoshop and alikes).

Regarding the "normalized gene expressions", and without knowing much about your specific experiment and data: I hope you are aware that such measures ([relative/normalized] expression) are usually not normal distributed, and that therefore "mean", "SEM" and "t-test" are usually not sensible. Often, it is reasonable to assume a log-normal distribution here, so you should do all your statistics on the log-expression values (where you can sensibly calculate means and SEMs and t-tests).

Also note that a bar plot is the (unfortunately most common but) worst possible way to present expression data. If your y-axis is "expression", note that the uncertainty is not symmetric around the location measure, so showing only one-sided error bars would be wrong (or at best incomplete). In any case, make sure that the bars extend to a meaningfur reference value (this may or may not be 1 for "expression" and 0 for "log expression", depends on the normalization and meaning of the data). A far better alternative is to show dots instead of bars (either a single dot/symbol at the position of the location measure or showing the complete data with each dot representing an individual measurement. You can then still use error bars to indicate SEMs or (better!) confidence intervals. Wheras bar plot highlight the absolute (mean) values compared to the reference (where often simply 0 is used, as in the attached example), the dot plots highlight the differences between the groups, what is typically the main research question. If you have >20 or so values per group you could also go for a box(-and-whisker-)plot (unfortunately not really available in Excel; workarounds and tricks known to me fail when values are negative). I attached an example showing the same data as "1d scatterplot" (individual data), "dot plot" (with error bars) and "box plot".

just Manually, insert SHAPE, and text in tab (format),

I'm more concerned about why you want to show the pvalues. Firstly, your sample sizes are painfully small so hopefully you've adequately addressed the required normality assumption for a t-test to work. Secondly, to find your pvalues that fall below the magical 0.05 you've likely calculated many pvalues subjecting yourself to a very high chance that your statistical signficiance is just a fluke (aka type I error). And Thirdly, you must address if the difference observed in the heart weight/tibia length ratio is clincally meaningful or not. Don't fall into the trap of getting all excited just becuase p < 0.05. Always remember the probability of a false positive is lingering in the background, waiting to rear its ugly head.

IF you are running a series of t-tests on more than 2 groups, you run into some other issues. Each time you compare 2 groups, you increase the chance of committing a type 1 error. You might be 95% certain any one test is good. But, if you have say 10 comparisons, you will only be about [(0.95)^10]% (59.87%) certain each of your tests are correct for type 1 errors.

I would use ANOVA analysis in your experiment and set your first group as a control group. Then use the Dunnet's test.

Andrew, just a side note: you are correct that the FWER increases. But still one must think is controlling the FWER is desired. It depends on the aim of the experiment (in screeningsfor single "candidates", I'd say, it is worth to control the FWER). And an ANOVA is nothing required to control the FWER. Dunnett's test will do, as well as others (Tukey, Bonferroni, Holm, ...). Almost all of these multiple-comparison.procedures control the FWER and do not need an ANOVA at all. There are only few, very special and strange exceptions that would need a so-called "ANOVA-protection" (e.g. Fisher's HSD for 3 groups).

Jochen,

I am referring to the 95% confidence level for any one test and about a 59.87% confidence level for all the comparisons.

Sorry for the confusion.

Andrew, I think I wasn't confused. All fine. I just wanted to stress the point that one should think about the meaning of "error rates", the differences between the different kinds of rates (EWER, FWER, FDR) and which of these reasonably should be controlled (and, actually, at what level?) in light of the study design and the research question(s).

you can use Daniel’s XL Toolbox, an excel add-on which have The Annotate Chart function

The Annotate Chart function provides a simple way to add comments and color to individual data points in your chart. For example, you can easily highlight specific points in a scatter plot, or you could add asterisks (“stars”, “*”) to a bar graph with a mouse click to denote statistical significance. 

http://xltoolbox.sourceforge.net/annotate.html#

Is it possible to do it with STATA?

Another excel add-on had the Annotate Chart function The XY Chart Labeler

http://www.appspro.com/Utilities/ChartLabeler.htm

http://www.appspro.com/Downloads/XYChartLabelerInstructions.pdf

 Sabry Abdallah are these pages available in English

I haven't figured out how to do the connecting lines in Excel, but I would strongly recommend that you use GraphPad Prism or SigmaPlot for this purpose.

Attached, however, is a template which you can use in Excel.

The code has been cracked after watching a YouTube video (https://www.youtube.com/watch?v=p1oy_opI8BY). Fairly simple, once you understand the principles governing the selection and formatting of chart elements.

The attached file contains a working example.

How can I insert statistical significance (i.e. t test P value < 0.05) annotations on top of my column bars on excel? - ResearchGate. Available from: https://www.researchgate.net/post/How_can_I_insert_statistical_significance_ie_t_test_P_value_005_annotations_on_top_of_my_column_bars_on_excel#589b634e3d7f4b2cc44cae93 [accessed Feb 8, 2017].

herewith the updated file

Below is a useful link for plotting significane in excel.

https://www.youtube.com/watch?v=p1oy_opI8BY

You can put everything you need .just click on chart tool>layout>insert text box.you can draw and design lines and put number or symbol into text box

Hi Paolo,

You can use BioVinci to do the ANOVA test for more than two groups of values. It automatically adds an annotation if there is a significantly different pair in asterisks (*).

Then use the 'edit plot' function to replace the asterisks with the annotations of p-value if you want to.

Hi, I support the contributors, good luck.

You can use Prism software in this context.

I do not think by using a statistical programs, might even be using photo editing softwares

Hello all!

I would apreciate a lot if you can advise me what would be the best way to show statistical significance when you have a one year study and a lot of data like one you can see on the attached image? I am already using Graphpad prism.

Thank you in advance

Hi Maja Diljkan

What significance do you want to show? differences between months ?, differences in the month?

Thank you Carlos Jimenez-Gallardo for assistance!

i`m just glad to help you