Take a step up from Excel

We hear it many times from our new customers – ‘Plots are ugly in Excel!’. You can always tell an Excel plot because it has very crude lines, a limited set of symbols and a gray plot background.

Isn’t it time to take a step up from Excel and create graphs you are proud to place in journals and on posters? Try KaleidaGraph today and find out how easy and inexpensive it is to create a publication quality scientific graph.

Make the switch today!

Here are just a few comments from our customers as to why they switched from Excel to KaleidaGraph:

• “The option of symbols are too limited in Excel.”
• “Graphs from KaleidaGraph exported as PICTs are able to be opened in FreeHand and imported as vectors, so I can do anything with the images – great for generating figures for publication. Excel is not even intended for this so I suggest you don’t even try.”
• “With Excel I was frustrated about the strict nature of the formatting and color choices.”
• “KaleidaGraph provides me with the ease of changing and saving preferred display styles. I have yet to get Excel to save a particular layout and style.”
• “Excel doesn’t do curve fitting very well, especially if you want to do non-linear curve fits (which it doesn’t do at all.)”
• “I was interested in making radar or spider plots and Excel could only make scatter diagrams.”
• “Excel lacks the ability to create multiple panels or multiple offset axes in one graph.”
• “Excel is not designed to perform in-depth scientific data analyses (statistical tests, i.e ANOVA).”

Friends don’t let friends use Excel for graphing and statistics

Jonathan D. Cryer from the University of Iowa statistics department states the following in an online article regarding the problems with Excel:

Good graphs should:

• portray numerical information visually without distortion
• contain no distracting elements (chartjunk)
• label axes (scales) and tick marks appropriately
• have a descriptive title and/or caption and legend

However, Excel meets virtually none of these criteria.

Scatter plots

Scatter plots represent bread-and-butter graphs for visualizing relationships between variables. Scatter plots should have:

• good choice of axes
• meaningful legends
• no false third dimensions

However, Excel’s default scatter plots leave much to be desired.

Histograms

Histograms are another basic statistical display. Histograms should have:

• no meaningless gaps
• a reasonable choice of bins
• an easy way to choose or adjust the bins
• a good aspect ratio
• meaningful labels on axes
• appropriate labels on bin tick marks

However, in Excel, the bin labels are impossible to read, the aspect ratio is poor and the legend and horizontal axis labels are useless.

‘Help’ Function

In any software, the help screens should give useful and accurate information, in particular, help screens should:

• not confuse
• give accurate statistical information
• be helpful!

However, Excel’s help for statistics is quite poor.

Treatment of missing data

Excel is especially deficient in its statistical analysis when some of the data are missing.

• Excel does it incorrectly
• Excel does it inconsistently
• Excel makes selecting predictor variables in regression especially difficult when data is missing

Computing algorithms for basic statistics

• Excel uses poor algorithms to find the standard deviation
• Excel defines the first quartile to be the ordered observation at position (n+3)/4
• Excel does not treat bed observations correctly when ranking
• Regression computations are often erroneous due to poor algorithms

In addition, Excel usually displays many more digits than appropriate.

Regression in Excel

Finally, Excel has major and documented difficulties with its regression procedures.

• does not treat zero-intercept models correctly
• sometimes gets negative sums of squares
• does not handle multicolinearly correctly
• computes standardized residuals incorrectly
• displays normal probability plots that are completely wrong
• makes variable selection very difficult

In summary, due to substantial deficiencies. Excel should not be used for statistical analysis. Get the right tool for the job. Try KaleidaGraph today!