"Frequently" Asked Questions

You guys claim this is a presentation tool, but all I see is a way to manipulate and chart data. What's wrong with you?
Um, the question itself is a bit too broad. Nonetheless, the general point is well taken. Our ultimate goal is to provide everything you need to make a presentation that embraces all the various kinds of evidence you need to make your point. But we had to start somewhere. Because a primary goal of presentations is (or should be) to guide decision-making based on concrete data, we decided that we would start with charting. We want to make it easy for you to analyze and display your data in order to make your point. If our tools are good enough, you may see something new.
How do I ...?
Depending on what you want, you may not be able to do it yet. Here's a partial list of features we are planning to build in the coming weeks. If you see something on here that would be very useful to you OR if what you need isn't listed below, please let us know so we can try to prioritize it.
  • Data
    1. Better data manipulation
      • combine or join
      • more powerful filtering
      • basic statistical analysis
      • creating "buckets" of data (taking a daily time series and reporting on monthly intervals)
    2. More data import formats
    3. More built-in data types
      • wider range of dates (e.g., just Years or just Years/Months)
      • currency (with inflation adjustment, at least for dollars)
      • latitude/longitude (GIS)
      • different kinds of dates
  • Charts
    1. Small Multiples for Charts
    2. More chart types
      • financial charts
      • maps
    3. Set X and Y axis to zero-origin independently
    4. Rating, commenting on charts
  • Presentations
    1. Create, edit, and share them.
  • Usability
    1. Screencasts on how to use features.
    2. Search
    3. More better, more grammaticaler help
    4. Nicer page titles
What is wrong with the existing tools like PowerPoint? Why should I use yet another tool for making presentations? But more importantly, why should I create presentations differently?
Rather than read our ramblings, we thought we'd let the experts speak on this one. Click here for an article from the New York Times by John Schwartz on exactly this question.
How do I upload data?
You first need to register and log in.

At the moment, there are three ways to upload data:

  1. Upload data in various formats (CSV, Excel/XLS, SPSS, SAS, Foxpro, etc.)
  2. Type or paste CSV data
  3. Grab it from a Google Spreadsheet (assuming you have an account there)
We plan on adding support for more web-based data formats (URLs of files, scraping data from tables on web pages, etc.). If you upload data that would be of interest to others, please click the "Allow others to View" button. And make sure you fill out all the fields since people will be far more likely to trust your data if you include a source reference and some description information
Hey, that Google Mashup is cool, but you guys suck, because all the spaces got stripped out of my carefully chosen column names. Why do you suck so much?
We blame Google (for this specific problem, not our general suckiness). The default column names they return are downcased and stripped of spaces and non-alphabetical characters (i.e., "# of Ways You Suck" becomes "ofwaysyousuck"). Fear not, we will fix this in the near future. In the meantime, you can edit the data and change the column names back to something sensible.
When I am looking at my data, I see a check box for "Allow Downloads?". What does this mean?
There are different levels of privacy for your data. If your data is private, which is how it starts out, no one can see, chart, or download it. If you make your data public, other people can see and chart it. If you "Allow Downloads" then they can also copy the data to their own computer.

(Note, if your data is public, particularly determined users could still get a copy of your data even if you don't "Allow Downloads". But it's not easy.)

Which type of chart should I use?
It depends on what kind of data you have. Here are some examples:
  • If you have time series data (e.g., inflation rates every month for the last 10 years), you probably want to use a line chart, with the X axis showing time and the Y axis the changing variable or variables.
  • If you want to compare the GDP of a country with its population size, you would use points (also known as a scatter plot) chart.
  • If you are comparing a number of countries' GDP, you would use a bar chart.
Which origin position is right for me and my family?
Most of the time, you should start the X and Y axes at zero. If not, you can easily play tricks with your data, creating the illusion of large variation. For example, if you are looking at a stock price and it fluctuates between 500 and 520, that's only a 4% difference. That is not that significant. But if you change the axis to have a range of 490-530, it'll look like a huge variation. But it's a lie. A lie which makes babies cry.
How should I pick my horizontal and vertical axes?
With a points chart, it's up to you to see which is easier to read.
With a line chart, if you are showing how a value has changed over time (e.g., the inflation rate in the last 10 years), the time value would go on the X or horizontal axis, and the Y axis would have the inflation rate. In general, show the variable whose change you want to look at on the Y axis. The X axis should be based on something that's evenly spaced (such as Time).
What does it mean to "group" data?
Here are two ways to represent the same data:

1.

DateWhoozit PriceWidget Price
1/1/05612
1/1/06713
1/1/071214
1/1/08102

2.

DateProductPrice
1/1/05Whoozit6
1/1/06Whoozit7
1/1/07Whoozit12
1/1/08Whoozit10
1/1/05Widget12
1/1/06Widget13
1/1/07Widget14
1/1/08Widget2

You want to use a line chart for this data.

In the first case, you would chart it by putting Date on the X axis and having two columns (Whoozit Price and Widget Price) on the Y axis.

In the second case, you would also put Date on the X axis. However, for the Y axis, you would chart Price and then "Group By" the Product column. This would create two groups, Whoozit and Widget, and you end up with the same two lines.

Same information, two different data representations, and so two different approaches to charting.

How do bubbles work?
If you are making a scatter plot, you can vary the size of the points to show another element of the data. Say you had this data:
CountryLife ExpectancyGDP Per Capita ($)Percent in Poverty
Fictionalstan60603270
Richville791603212
Upandcomingcountry661003235
You could chart with Life Expectancy on the X axis, GDP on the Y Axis, and vary the size of the points based on Percent in Poverty:
Faq_dot_size
How do chart colors work?
They are similar to Bubbles, but instead of varying the size of each point based on a column of data, the color is changed. For line charts, it changes each line segment.
What's the difference between displaying the Y-axis normally with a line chart and transforming the values to be relative to the first value?
If you are just plotting a single column, there isn't much purpose to it. If you are grouping your data or plotting multiple columns, it allows you to compare the changes between data. For example, if you are comparing stock data and just plot the raw values, you get this:

Faq_transformation_1

From this chart, it's hard to see which stocks performed best. At a glance, it seems like IBM and AAPL are superior to the rest. But if we plot them as relative values, we get a different picture (literally and figuratively):

Faq_transformation_2

Now it's clear that the only breakout stock is AAPL (which more than doubled), with the rest performing roughly the same.

I uploaded data with amazing precision. I have numbers like "3.14159265258979323" and your system is truncating them to "3.14159265258979". Why? Why? Why?!?!?!?!?!?!
As you may know, floating point representation on computers is a tricky, imprecise business. At the moment, in order to help things run faster, we are limiting the precision of floating point numbers. If you have data for which this is a problem, please let us know.
Why do use the word "average" everywhere you really mean "arithmetic mean?" Don't you know that mean and median are both averages?
Yeah, you got us. Unless it's been defaced, Wikipedia has a cogent explanation of this. However, we're trying to walk the line between clarity and annoying. If you know the difference, you'll know what we mean (but perhaps you'll be a little annoyed). If you don't know the difference, you'll still know what we mean. Either way, we've made our point.
Don't you know that the word "data" is the plural of "datum?" Why don't you write, "The data show this" as opposed to "The data shows this?"
Hey, did you ask the last question? Regardless, most people think of data as singular. It strikes us a vaguely pretentious to use it as a plural. (Except when you do it. When you do it, it just sounds smart. But from us, it sounds pretentious. Trust us.) The Wiktionary agrees with us. And if you're going to be pedantic, we trust that when you plan a meeting, your agenda contain many items, because agenda is the plural of agendum.
I didn't get a data point annotation line. But I clicked the "T" symbol and my cursor became a "T" and I clicked on the charted line?
I bet, if you choose "points and lines" as your charting type, that you will see that the point that you wanted to annotate wasn't actually a data point; it was probably the part of the line that connected two points. Currently we only allow annotation of "actual data points". Sorry. We are planning sometime on having annotation of "arbitrary" points on the chart with lines to the arbitrary points. If this is a feature that you would like, please drop us a line.
My title is kind of long and doesn't fit on the title line. Are you guy's Bill Gate's-ing me on how big titles should be?
No we are trying desperately to avoid the whole "No one will ever need more than 32K of RAM" trap. In our world, a title is a single line. If your's doesn't fit, you can change the font size or you can "hide" the title and open a multi-line text box in it's place.
How do I hide/show my Title,Description,X Axis label,Y Axis label,Transform description, or Legend?
Bring up the "Chart Annotations" modal window by going to Chart->Annotations. Any one of the afore-mentioned text strings will be displayed if the check box is checked or hidden if it is NOT checked.
How do I create a multi-line RTF text box to annotate my chart?
Click Chart->Add Text. Your cursor will become a capital T. Click on the spot in your chart where you want to place your text box. Double click in the box and type away. You can also get into Text Mode (where your cursor becomes a capital T) by clicking on the T "Text" icon.
How do I annotate a particular data point with a descriptive RTF text box?
Click Chart->Add Text. Your cursor will become a capital T. Click on the exact Data Point in your chart that you want to annotate. Double click in the box and type away. You will notice that a line connecting your text box to that data point will be created allowing you to move the text box anywhere on the chart and still maintain the connection to that one specific data point. You can also get into Text Mode (where your cursor becomes a capital T) by clicking on the T "Text" icon.
I sense a disturbance in the force. Something changed since the time I was here. What is it?
Great question. You should check out our list of changes to find out.
Your featured data sets all have pretty icons. How do I get a pretty icon for my data. Who has the copyright on the icons you are displaying?
We will make it easy in the near future to let you pick a public Flickr photo for inclusion in your data set. We're using Creative Commons licensed photos right now. Here's a list of the ones we are using on the site and where to find them on Flickr. (We are also using public domain photos we found on Flickr that we don't list below.)