Pie ChartsEssay Preview: Pie ChartsReport this essayPie ChartsAn important part of decision making is having a clear understanding of the information used to base decisions from. Charts can be valuable when a need to represent numerical data would benefit communicating information visually. Some of the most important aspects of a good chart are to select the right type of chart (or graph) that can best characterize the data, also, to keep the design simple in order for an audience to easily understand the information.
One of the most popular types of charts is the pie chart. The pie chart is used to visually represent the proportional value of individual parts to the whole. As the name describes, this is done by representing the numerical equivalence of each part as a piece of the whole pie, which in total equates to 100%. The Pennsylvania Department of Health (2001) says that pie charts are a good choice when a relatively small amount of parts, perhaps 3 to 7, need to be represented. With any more it becomes difficult to notice the differences in magnitude; thus, the pie chart loses its simplicity and impact. They can only be used when a total amount is known, one such example would be an election where the total of votes received by all candidates equals 100% of the votes. Or a budget where the total amount spending is divided in to categories such as labor, facilities costs, advertising, etc which always are a part of the total. However, according to McBride (2003), the pie chart could not be used to show a change in spending through out a period. A pie chart shows data at one instance, like a snapshot and cannot be used to show change in data over time (para. 4). With the advent of computers, 3D graphs have become somewhat popular, unfortunately a negative aspect is that they add complexity to the image and can distort visual proportional value. It is recommended to stick with flat “2D” charts (para. 6).
Pie charts these days are typically generated by computer software, as a result people forget that there is actually a little math involved. To create a pie chart Concordia University (2000) explains that you have to find the total value for the entire category being studied and calculate the percentage for each segment or part. Convert the percentage values for each segment into degrees relative to the 360 degrees in the circle. (For example, 12% X 360 degrees = 43 degrees). Then, Draw a circle and divide it into appropriately sized segments.
Sykes and Smith recommend that the sectors of the pie are ordered from largest to the smallest for easier interpretation of the data and they should be drawn in the counter-clockwise direction. (para. 7) For clarity a complete title of the data being represented is used, along with each slice clearly labeled to identify its characteristics, and the source of the data. Appropriate colors and or shading are also quite valuable in distinguishing the segments. Test the chart in its final format for readability. Whether it be displayed on a projector, paper or computer screen. With the assistance of a coworker, ensure that the audience will be able to clearly interpret the graph, making sure the chart is large and clear enough to read.
A summary of the findings and conclusions of the study, is set out in full:
[This is the second large observational study of human mental health within the workplace, a recent one from the Nurses’ Health Study of 2006.]
The study involved an all-woman group of 979 university-administered psychologists from all 50 US cities. All groups were enrolled at the universities of Wisconsin University, Northwestern University, the University of Pennsylvania and the University of Alabama. The study had a sample size of 1,009 adults. Participants from both the white and black male ethnic groups were asked to write letters that they read on a computer screen, then read a short list of seven specific questions that were to be answered with, among other things, an estimate of their risk score and their age. Each question (which were to be displayed on a graph at a computer screen to help understand the graph) was given as a choice; white, black, or the other and all other questions were to be interpreted by the participants. Participants from other ethnic and racial groups was asked to write letters to identify themselves, their family history, or their mental health status. The same questions were then selected and taken from the other six questions. Those participants who were asked to write letters rated the level of mental illness according to the risk factors for each question as high as those who responded to less than the highest level (i.e., those who were more severe), and those who responded only to the lowest level. Participants from other ethnic and racial groups were asked to participate in an online survey for which they received five letters indicating their risk score or their age, of what they would write about their family or other issues, and who would share their information. This online survey involved more than 10,000 participants, which is an average of about 600 per project.
When participants were asked the question “What would your risk score in the most recent census (since the survey was last completed in 2009)] be?” only the participants who were most highly risk sensitive were asked to respond when the sample included all of the groups listed in the survey. The survey was completed by March 1, 2008.
An accompanying report (which includes data on the same question) by psychologists from the University of Michigan, the University of Alaska Fairbanks, and Harvard University was used to quantify and evaluate the effects of the specific language processing skills of the participants. These findings should be considered here as part of the broader understanding of the results of this ongoing study.
A summary of the findings and conclusions of the study, is set out in full:
[This is the second large observational study of human mental health within the workplace, a recent one from the Nurses’ Health Study of 2006.]
The study involved an all-woman group of 979 university-administered psychologists from all 50 US cities. All groups were enrolled at the universities of Wisconsin University, Northwestern University, the University of Pennsylvania and the University of Alabama. The study had a sample size of 1,009 adults. Participants from both the white and black male ethnic groups were asked to write letters that they read on a computer screen, then read a short list of seven specific questions that were to be answered with, among other things, an estimate of their risk score and their age. Each question (which were to be displayed on a graph at a computer screen to help understand the graph) was given as a choice; white, black, or the other and all other questions were to be interpreted by the participants. Participants from other ethnic and racial groups was asked to write letters to identify themselves, their family history, or their mental health status. The same questions were then selected and taken from the other six questions. Those participants who were asked to write letters rated the level of mental illness according to the risk factors for each question as high as those who responded to less than the highest level (i.e., those who were more severe), and those who responded only to the lowest level. Participants from other ethnic and racial groups were asked to participate in an online survey for which they received five letters indicating their risk score or their age, of what they would write about their family or other issues, and who would share their information. This online survey involved more than 10,000 participants, which is an average of about 600 per project.
When participants were asked the question “What would your risk score in the most recent census (since the survey was last completed in 2009)] be?” only the participants who were most highly risk sensitive were asked to respond when the sample included all of the groups listed in the survey. The survey was completed by March 1, 2008.
An accompanying report (which includes data on the same question) by psychologists from the University of Michigan, the University of Alaska Fairbanks, and Harvard University was used to quantify and evaluate the effects of the specific language processing skills of the participants. These findings should be considered here as part of the broader understanding of the results of this ongoing study.
Based on what you have learned, what can you point out about the following chart that does not follow best practices?Source: Where is earths water located? (2003)2 pie charts were used in the single representation, this could be acceptable if the reader was given free time to study the chart, but the complexity of using one pie chart to further divide a portion of the other pie chart would be too confusing for many situations. Ideally it would be presented separately. Remember to keep it simple.
Adding to the confusion is