Figures and Tables Throughout the course, you will be presenting your data graphically.

Usually, this is done in the form of a figure or table. In this class these will be generated using Microsoft Excel. If you generally have used Microsoft Office 97, 2003, XP, etc. and have recently acquired Vista, you have likely noticed that there are many differences in these operating systems. The computers in the PMBA Lab have the older version of Office but Vista is equally adept at creating graphs. If you have any questions about how to generate graphs using either system ask your TA. There are specific rules that come along with both of these. One major rule is that you should almost never display the same data in both figure and table formchoose which one is best. Figures are most often graphs that include an x- and y-axis (and possibly a z-axis if being presented in three dimensions), although they could be a different form, such as a pie chart. In this class, you will usually create column charts or XY scatterplots. The type of graph utilized depends on the information being presented, as well as the experimental design and statistics used. Other types of figures you may see include photos, drawings, and phylogenetic trees. A caption is included at the bottom of the figure, and this caption has a specific format. An example of this is given on the next page. Tables are listings of data. This may include raw data (rarely), lists of calculated data such as means and standard deviations, groupings of P-values, etc. Tables also always contain a caption, but table captions are found at the top. Neither figures nor tables have titles. All tables and figures should stand on their own. Anyone who may find them, without the supporting data from your report or abstract, should be able to interpret them. Experiment: The following is a completely made-up experiment and data set. The researchers were interested in seeing whether diet could affect growth rate in mice, so they set up three different diets. In the first diet, mice were fed a specific amount of powdered food daily. This amount was 80% of that found to be normally consumed by mice in a previous study. In the second diet, mice were allowed to eat as much as they wanted, whenever they wanted (this is called ad libitum feeding and would correspond to 100% of the amount eaten in the previous study). The third diet was also ad libitum, but the food was spiked with growth hormone. All animals were housed separately and were allowed access to water at all times. Animals started on their experimental diet 7 days after birth, at which time they were able to eat solid food. Every 7 days after commencement of the experimental diet, mice were weighed individually. Table 1 shows the raw data generated during the study. Figure 1 is a graphical comparison between the three groups at each time point.

Table 1. Raw feeding trial mass (g) data. C57Bl6++ mice (N = 30) were fed one of three experimental diets. Mice were housed individually with access to water at all times and were weighed weekly. Means and standard deviations are given for each treatment at each time. Treatment Restricted Animal # 1 2 3 4 5 6 7 8 9 10 14 4.8 5.2 4.9 5.0 5.5 4.7 5.1 5.1 4.4 4.6 4.93 0.320 6.2 6.5 5.8 5.5 5.7 6.0 5.6 6.2 4.9 5.5 5.79 0.458 8.2 8.7 9.0 8.8 7.8 7.9 8.3 8.5 8.3 7.8 8.33 0.422 Age (Days) 21 28 8.3 12.4 8.6 13.3 9.1 12.5 8.2 12.8 7.8 13.1 7.7 12.0 8.5 12.5 9.2 13.4 8.8 12.9 8.5 12.7 8.47 12.8 0.495 0.433 12.7 13.1 13.8 13.4 12.6 14.1 13.5 12.5 12.8 12.6 13.1 0.563 16.7 17.2 15.4 15.6 16.3 17.8 16.6 16.7 15.8 16.2 16.4 0.735 18.9 19.4 19.9 18.7 21.0 19.4 19.4 19.7 18.5 18.8 19.4 0.730 22.4 23.6 22.8 23.1 24.1 25.2 24.6 22.9 23.7 24.1 23.7 0.873

Mean SD Ad Libitum 11 12 13 14 15 16 17 18 19 20

35 16.7 17.1 16.3 17.5 17.2 16.9 16.2 17.4 17.8 18.2 17.1 0.633 24.3 25.1 23.6 24.7 25.8 24.5 25.2 24.9 25.5 25.3 24.9 0.642 30.2 31.5 29.8 30.5 31.3 31.1 29.9 30.4 30.8 30.7 30.6 0.571

42 24.6 25.7 23.6 25.4 24.7 24.3 23.8 24.6 23.8 25.3 24.6 0.724 31.2 30.8 31.1 32.4 30.5 32.1 31.8 31.5 31.4 32.5 31.5 0.667 36.5 35.8 35.7 36.2 36.5 34.9 35.2 35.7 35.9 35.0 35.7 0.572

Mean SD Growth Hormone 21 22 23 24 25 26 27 28 29 30

Mean SD

40 35 30 Mass (g) 25 20

c c c c b a a b a b

Restricted Ad Lib Growth Hormone

15 10 5 0 14

b b c a

21

28 Age (d)

35

42

Figure 1. Growth rates in mice consuming different diets. C57Bl6++ mice ( M us musculus ; N = 30 ) were fed on one of three dietary regimens indicated and were weighed ev week. Animals had access to water at all times. Error bars indicate +/- 1 ery SD, and letters indicate statistically significant groupings at P < 0.05 using a one-way ANOVA with Tukey post-hoc test for each age group. Figure 1. Fragment describing what is shown in the figure. Brief description of the important aspects of the experimental methods (this doesn't hav to be v detailed, but e ery should cov important points such as species, N, basic experimental design, er Indication of what error bars, letters, asterisks, etc . mean. Statistical tests used.

etc. ).

Notes: 1) Get rid of lines and background colors on graphs. Likewise, use only shades of black and white or print in grayscale to save ink (unless using your own printer). 2) Make sure your caption is unique (i.e. do not copy this one). 3) Make certain that you include units where needed. 4) Do not use abbreviations without defining them, unless they are standard. 5a) Error bars in Microsoft Office 97, XP, etc. are added by right clicking on a column and selecting Format Data Series. Select the tab labeled Y Error Bars and click on the circle for Custom. At that point, you need to enter the cells for your standard deviation data in both the + and - boxes. 5b) Error bars in Microsoft Office Vista are added by clicking on a column and selecting Error Bars in the Analysis Box under the Layout tab in the Chart Tools Area. Select the More Error Bars Options At that point, select the Custom bullet point and click the box labeled Specify Value. Then select the cells or cells (for multiple columns/means) that have your standard deviations for the Positive Error Value and Negative Error Value boxes. 6) Add a textbox below the figure for your caption and above bars for * and letters. 7) Check your significant figures to ensure consistency.

8) Conserve paper and ink by eliminating shading and gridlines. 9) For one comparison, an asterisk (*) indicates significant difference; for multiple comparisons, letters indicate significant groupings (i.e. those with the same letter are not different from each other statistically) 10) Try to fit more than one figure per page (you do not need to have full-page figures). 11) Right justify numbers in table columns. Common Problems on Figures: - Saludos estudiantes seccin 101 y 102: ATENCION ESTO ES IMPORTANTE, att: REY ROSA 1) 2) 3) 4) 5) 6) 7) 8) 9) 10) 11) 12) 13) 14) 15) 16) 17) 18) 19) Axis with different decimal places (e.g. 0, 0.05, 0.1, 0.15). Unnecessary decimals (e.g. 0.00, 10.00, 20.00). Improper labeling of axes (especially column figures). No units on axes. Improper error bars (not standard deviation, same error bars for each column). Missing error bars. No explanation of error bars. No indication of significance (* or letters). Improper indication of significance. No explanation of significance in caption (e.g. there was a sig. diff. isnt good). No Figure 1 at beginning of caption or Figure 1 not bolded. Sentence, rather than fragment at beginning of caption. Restatement of axes, rather than unique description in caption fragment. No indication of methods. No indication of sample size (N) No explanation of statistics. Improper/incomplete explanation of statistics. Poor or awkward grammar in caption. Anything mentioned previously.

Reporting P Values 1) Proper P-value format (should be P < 0.01, not p<.01). 2) If P < 0.01, report as P < 0.01 or P << 0.01 or P < 0.001 3) If P > 0.01, report actual P value to 3 decimal places (e.g. P = 0.012 or P = 0.546) 4) Dont state as P > 0.05 or P < 0.05 unless performing an ANOVA 5) P values should always be reported in captions, but they can be parenthetically 6) Dont make a P value the subject of a sentence, it is only a footnote 7) When doing a regression analysis, always report the P value below the R2 value on the figure Parentheses When writing a caption, report, or abstract, using parentheses to state information about the experiment can be very useful. Since the focus of your statement should be the result or methods of the experiment, things like sample size, species, stats, P-values, etc. can be stated parenthetically to save space. For example, in the sentence The mice fed a diet spiked with growth hormone weighed significantly more than the mice fed a restricted diet after 21 days (P < 0.001, one-tailed, unpaired t-test) the result is the main focus of the sentence, but the P-value and stats are also reported.