Teaching talk:TUW - UE InfoVis WS 2009/10 - Gruppe 09 - Aufgabe 2

Bewertung

endgültige Bewertung: 20 Punkte

Begründung:

• Alle Änderungen durchgeführt.

Katharina-Anna Wendelin 21:18, 21 Dezember 2009

---

vorläufige Bewertung: 19 Punkte

Begründung:

• Vorsicht : Dezimalstellen sollen einen Punkt haben, Tausenderstellen ein Komma.
• Die white Spaces in Tabelle 2 (zwischen den Spalten) sind nicht einheitlich.
• Die vertikalen Spanner-Header in beiden Tabellen sollten zentriert sein.
• Sehr gute Diskussionsseite ;)

Katharina-Anna Wendelin 22:09, 30. November 2009

---

--UE-InfoVis0910 0625847 15:20, 15 November 2009 (CET)

Assumption
I have been looking on http://sti.srs.gov/fulltext/tr2002506/tr2002506.html to learn more about the topic of this table. Decontaminated liquid waste has to be disposed! Well, samples of solids were removed from Tank 50H during October of 2002 and sent to the Savannah River Technology Center for analysis [Wilmarth et al., 2002]. Three core samples were sent to SRTC (labeled as TK50-HTF-E-195, TK50-HTF-E-196, and TK50-HTF-E-197). So our table contains the results from radiochemical analysis of three samples from Tank 50H using Aqua Regia Digestion. And the samples were analyzed in duplicate.

Inprovement propositions

White Space
White space can be intentionally manipulated to direct your readers’ eyes to either scan predominantly across the columns or down the rows [Few, 2004]. So I would prefer to direct the readers to scan predominantly across the rows so that comparision between the three samples in view of a certain element could be done easier. Hence the the vertical white space between the rows should be bigger. But be careful: This balance between white space and overall data density is upset when the vertical white space between the rows exceeds the vertical space used by the rows of data themselves [Few, 2004].

Rules and Grids
Grids are useless in view of delineating columns and rows [Few, 2004]. Als rules have only a limited usefulness. The problem with rules and grids is that they break up the data (the lines distract the eye, promoting a strong perception of individual cells through the Gestalt principle of enclosure rather than a seamless flow of information). So I would delete the whole grids. Furthermore I would only use rules to separate the header and footer from the body.

Columns and rows
You should almost always arrange the categorical subdivisions bidirectionally when you can fit one or more sets of categorical subdivisions across the columns [Few, 2004]. So I wouldn’t do any changes.

Data Sequence Alignment
Numbers that represent quantitative values, as opposed to those that are merely identifiers (e.g., customer numbers), should always be aligned to the right [Few, 2004] So I would do so instead of aligning it to the centre. Furthermore I would align both the decimal point and the final digit to the right. This can be accomplished by expressing each value using the same number of decimal digits, even when they are zeroes [Few., 2004]. Now lets have a look at the text. Text that expresses neither numbers nor dates works best when aligned to the left because of the historical conventions of printing [Few, 2004]. So changing into this would be adviseable.
Furthermore the headers should be aligned like their associated data, because this is international [Few, 2004]. E.g.: ff the column’s data are left aligned, its header is left aligned as well, and so on.

Number Precision
The level of precision should not exceed the level needed to serve your communication objectives and the needs of your readers [Few, 2004].

Measuring Units
Columns should be titled with the name of the variables followed by the units of measure in parentheses[Wallace, 2005]. I woud do so and delete the column called unit.

Spanner Headers
We could use a spanner header per sample e.g.: TK50-HTK-E-195 for the first sample and 1 and 2 as columnheaders for the two columns below the spanner header which stands for the duplicate analysis of one sample.

Ambiguities

• I don’t know but are the units dgm/g and microg/g compatible or comparative? If not, then I would break the sets of data into smaller groups in view of their unit
• No americium-241 was detected in the solid samples [Wilmarth et al., 2002] . So if there is no americium-241 I think it should be announced in the table
• Furthermore the table contains the results from radiochemical analysis using gamma spectroscopy, alpha spectroscopy and Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) Few., 2004. Does this mean that for every sample another method (e.g.: gamma analysis) has been used? If so then I think we should note this fact in the table.

Refereces

[Few, 2004] Stephen Few. Show Me the Numbers: Designing Tables and Graphs to Enlighten. Analytics Press, 2004, Chapter 8 – Table Design

[Wallace, 2005] Rosa Wallace. Designing Tables. http://www.ncsu.edu/labwrite/res/gh/gh-tables.html

[Wilmarth et al., 2002] W. R. Wilmarth, C. J. Coleman, F. F. Fondeur, V. H. Dukes, M. P. Bussey, M. S. Blume, and A. V. Bowman. Results of Sample Analysis from Solids Removed from Tank 50H. http://sti.srs.gov/fulltext/tr2002506/tr2002506.html

---

Units

• dpm/g are disintegrations per minute per gram.
• microg/g are micrograms per gram.

I suggest adding footnotes describing the abbreviations.

The two units are not comparable, I suggest splitting the information into two separate tables. The first should contain data about Sb-125, Cs-137, Sr-90, Pu-239/240 and Pu-238, the second table data about the different Uranium isotopes.

Data Rows

• Sb is Antimony, Sb-125 is a synthetic radioisotope.
• Cs is Caesium, Cs-137 occurs naturally only in trace amounts.
• Sr is Strontium, Sr-90 occurs naturally only in trace amounts.
• Pu is Plutonium, Pu-239 occurs naturally only in trace amounts, Pu-238 and Pu-240 are synthetic radioisotopes.
• 232, 235, 236, 238 and 239 are different isotopes of Uranium, so I'd suggest naming them U-232, U-235, U-236, U-238 and U-239 respectively for consistency reasons. U-232 and U-239 are synthetic radioisotopes; U-236 occurs naturally only in trace amounts; U-235 has a natural abundance of 0,72%; U-238 of 99,27%.
• % U Enrichment measures the fraction of U-235 (which is fissile), the natural concentration of which is 0,71%.

Should we mark synthetic radioisotopes to distinguish them from naturally occuring isotopes?

Other

The percentage sign should be placed to the right of each value. [Few, 2004]

--UE-InfoVis0910 9930270 14:27, 16 November 2009 (CET)

UE-InfoVis0910 0625100 15:40, 16 November 2009 (CET)

I think the abbreviations are not necessary, because if you know what they mean you won't need a description.

UE-InfoVis0910 9930270 17:03, 17 November 2009 (CET)

I would recommend marking missing values (currently listed as being zero) using three dashes ("---"). We don't know if the specific analyte was not found in the sample, if no test was performed, or if it is below a certain detectable amount. Listing it as "0,00" probably adds wrong information to the table.