When to use tables - How to design effective tables - Preparing the tables and figures

How to write and publish a scientific paper - Barbara Gastel, Robert A. Day 2022

When to use tables
How to design effective tables
Preparing the tables and figures

A tabular presentation of data is often the heart or, better, the brain, of a scientific paper.

—Peter Morgan

When to use tables

Before proceeding to the “how to” of tables, let us first examine the question of “whether to.”

As a rule, do not construct a table unless repetitive data must be presented. There are two reasons for this general rule. First, it is simply not good science to regurgitate reams of data just because you have them in your laboratory notebooks; only samples and breakpoints need be given. Second, the cost of publishing tables can be high compared with that of text, and all of us involved with the generation and publication of scientific literature should worry about costs.

If you made (or need to present) only a few determinations, give the data in the text. Tables 16.1 and 16.2 are useless, and yet they are typical of many tables that are submitted to journals.

Table 16.1 is faulty because two of the columns give standard conditions, not variables and not data. If temperature is a variable in the experiments, it can have its own column. If all experiments were done at the same temperature, however, this single bit of information should be noted in the materials and methods section and perhaps as a footnote to the table, but not in a column in the table. The data presented in the table can be given in the text itself in a form that is readily comprehensible to the reader without taking up space with a table. Very simply, these results would read: “Aeration of the growth medium was essential for the growth of Streptomyces coelicolor. At room temperature (24°C), no growth was evident in stationary (unaerated) cultures, whereas substantial growth (OD, 78 Klett units) occurred in shaken cultures.”

Table 16.1. Effect of Aeration on Growth of Streptomyces coelicolor

Temp (°C)


No. of Expts.


Aeration of Growth Medium


Growtha

24


5


+b


78

24


5



0

aAs determined by optical density (Klett unit).

bSymbols: +, 500-ml Erlenmeyer flasks were aerated by having a graduate students blow into the bottles for 15 min out of each hour; −, identical test conditions, except that the aeration was provided by an elderly professor.

Table 16.2. Effect of Temperature on Growth of Oak (Quercus) Seedlingsa

Temp (°C)


Growth in 48 h (mm)

−50


0

−40


0

−30


0

−20


0

−10


0

0


0

10


0

20


7

30


8

40


1

50


0

60


0

70


0

80


0

90


0

100


0

aEach individual seedling was maintained in an individual round pot, 10 cm in diameter and 100 cm high, in a rich growth medium containing 50 percent Michigan peat and 50 percent dried horse manure. Actually, it wasn’t “50 percent Michigan”; the peat was 100 percent “Michigan,” all of it coming from that state. And the manure wasn’t half-dried (50 percent), it was all dried. And, come to think about it, I should have said “50 percent dried manure (horse)”; I didn’t dry the horse at all.

Table 16.2 has no columns of identical readings, and it looks like a good table. But is it? The independent variable column (temperature) looks reasonable enough, but the dependent variable column (growth) has a suspicious number of zeros. You should question any table with a large number of zeros (whatever the unit of measurement) or a large number of 100s when percentages are used. Table 16.2 is a useless table because all it tells us is that “The oak seedlings grew at temperatures between 20°C and 40°C; no measurable growth occurred at temperatures below 20°C or above 40°C.”

In addition to zeros and 100s, be suspicious of plus and minus signs. Table 16.3 is of a type that often appears in print, although it is obviously not very informative. All this table tells us is that “S. griseus, S. coelicolor, S. everycolor, and S. rainbowensky grew under aerobic conditions, whereas S. nocolor and S. greenicus required anaerobic conditions.” Whenever a table, or columns within a table, can be readily put into words, do it.

Table 16.3. Oxygen Requirement of Various Species of Streptomyces

Organism


Growth Under Aerobic Conditionsa


Growth Under anaerobic Conditions

Streptomyces griseus


+


Streptomyces coelicolor


+


Streptomyces nocolor



+

Streptomyces everycolor


+


Streptomyces greenicus



+

Streptomyces rainbowensky


+


aSee Table 16.1 for explanation of symbols. In this experiment, the cultures were aerated by a shaking machine (New Brunswick Shaking Co., Scientific, NJ).

Table 16.4. Bacteriological Failure Rates

Nocillin


K Penicillin

5/35 (14)a


9/34 (26)

aResults expressed as number of failures/total, which is then converted to a percentage (within parentheses). P = 0.21.

Some authors believe that all numerical data must be put in a table. Table 16.4 is a sad example of this phenomenon. It gets sadder when we learn (at the end of the footnote) that the results were not statistically significant anyway (P = 0.21). If these data were worth publishing (which seems doubtful), one sentence in the results would have done the job: “The difference between the failure rates—14 percent (5 of 35) for nocillin and 26 percent (9 of 34) for potassium penicillin V—was not significant (P = 0.21).”

In presenting numbers, give only significant figures. Nonsignificant figures may mislead the reader by creating a false sense of precision; they also make comparison of the data more difficult. Unessential data, such as laboratory numbers, results of simple calculations, and columns that show no significant variations, should be omitted.

Another very common but often useless table is the word list. Such a list might be suitable for a slide in a presentation, but it does not belong in a scientific paper. Table 16.5 is an example. This information could easily be presented in the text. A good copy editor will kill this kind of table and incorporate the data into the text. Yet, when copy editors do so (and this leads to the next rule about tables), they often find that much or all of the information was already in the text. Thus, the rule: Present the data in the text, or in a table, or in a figure. Never present the same data in more than one way. Of course, selected data from a table can be singled out for discussion in the text.

Table 16.5. Adverse Effects of Nicklecillin in 24 Adult Patients

No. of Patients


Side Effect

14


Diarrhea

5


Eosinophilia (≥5 eos/mm3)

2


Metallic tastea

1


Yeast vaginitisb

1


Mild rise in urea nitrogen

1


Hematuria (8—10 rbc/hpf)

aBoth of the patients who tasted metallic worked in a zinc mine.

bThe infecting organism was a rare strain of Candida albicans that causes vaginitis in yeasts, but not in humans.

Tables 16.1 to 16.5 provide typical examples of the kinds of material that should not be tabulated. Now let us look at material that should be tabulated.