DataAnalysis is a general purpose iPad App for the plotting and fitting of all types of data that can be formulated as x, y pairs. The program can be used easily by both students and professionals. It is particularly useful for quick analyses of various types of data by curve fitting, value prediction via a standard curve, and analysis of kinetic data including enzyme kinetics.
Data can be entered directly via the keyboard or imported from the Mail App as a text file (.txt), and a comma separated value file (.csv). In addition, data can be pasted from other Apps and imported via drag and drop from within iTunes. Data files in txt, csv and native formats can be imported directly from Dropbox.
After data entry the user has the option to average, baseline correct, and/or normalize the data before plotting and fitting.
Plotting of the data is as simple as pressing the Plot button. The axes are autoscaled and the graph is immediately ready to add to a document or presentation. The user has a variety of options to customize the graph by altering the axes, axes labels, graph title, and changing the symbol, its size and color.
By using touch and drag the user can create a selection box. Upon selection of the data subset, the user has two options. (1) To zoom in on the selected data. Any fitting done in this mode is global (e.g. uses all the data). (2) To zoom and perform a fitting only on the selected subset of the data
The data can be fit to a variety of mathematical equations by non-linear regression including simple functions such as linear, 2nd and 3th order polynomials, power, exponential, and log functions.
Data can also be fit to more complex equations for such processes as radioactive decay, enzyme kinetics, 1st and 2nd order chemical reactions, as well as gaussian, logistics and surge functions. The fit can be overlaid on the data with different line widths, types and thicknesses.
Lowess Regression (LOcal Weighted regrESSion) can also be used to fit data without the need to specify a mathematical function. Not only does Lowess Regression not require a mathematical function, but also does not require parameters (non-parametric). This fitting method is used extensively to fit everything from election polling to astrophysics data, where the data have a large degree of scatter.
For many analytical situations, the data fit can be used as a standard curve to determine the value of unknowns. This analytical procedure is completely automated within DataAnalysis.
The program output, graphics and text, can be copied, emailed, or uploaded to Dropbox in a variety of formats including PDF, PNG and TXT. The graphics in PDF format can be edited with programs like Adobe Illustrator™ and AutoDesk Graphic™ as objects.
DataAnalysis also supports multitasking and printing of input data as well as all forms of output.
DataAnalysis has built in tool tips, a comprehensive web based User Guide and how-to videos all available via the Help Menu.
Version 3.4.5 of the DataAnalysis App has a number of important bug fixes. In addition, DataAnalysis native format data files are now compatible with the new iPhone version call DataAnalysisLite.
Ratings and Reviews
Works as Described
The input and output are not very fancy but if you need a simple x-y plot, this is it. It produces a nice graph and will fit a polynomial line (via regression) of your choice together with relevant statistics.
My first use was to plot some blood sugar readings that I take daily. It did an excellent job.
I am giving it another star.
Extrapolation a pain
I Love the concept and understand the potential power, but my first data entry attempt included negative y values; which I am told is a non- numerical character in an error message! (UPDATE: I found a way to fixed this with a less intuitive data entry method)
Perhaps my biggest problem after playing with some spectrophotometry curves and enzyme velocity curves is that once I enter a set of data, plot it, then fit the appropriate curve, then try to use it as a standard curve; any data that is beyond the original data is off the graph and out of sight--the app has no automatic re-scaling function that I can find. My data frequently requires extrapolation and without a large enough area on the plot, I am limited to interpolation.
I wanted to use this with my students, and at my bench. Without extrapolation (& auto re-scaling)...no dice.
I wanted to contact the developer, but the developer website kept forcing me to the "manual" and provided no contact info.
Good. Only thing strange
This is potential very useful, but its usefulness quickly declined for me when I found what appears to be a major flaw in logic. When you plot x,y values, x is almost always the independent variable in any function. Therefore, when I fit a function to a set of data, what I would like to do is input an x value to see what the corresponding y value is. Not the other way round.
If I know that 1 apple cost 25 cents, 2 apples cost 42 cents and 5 apples cost 96 cent and I plot this data, I would like to know how much 3 apples cost. It does not make any sense for me to have to guess a price (y value) to see how may apple I can buy (x value). Although that may sometimes be useful, the y value is what I really want to know by entering the number of apples (x value).
If you can change this, you can charge for the app, seriously.
With Family Sharing set up, up to six family members can use this app.