Lay and professional vocabulary

All tags were put in categories according to how they can be characterized as lay, professional or neutral. The professional vocabulary is cancer medicine. The categorization is based on familiarity with Danish language and its use. No statistics of word usage among laypersons and health professionals were used. This is due to the fact that laypersons and health professionals tend to use each others’ vocabulary (see also chapter 8.3.5 and Glenton (2006)). It was a goal to categorize the tags based on a combination of their meaning, origin and general familiarity, more than their use.

Five categories were used:

0,0 20,0 40,0 60,0

nov…

des.…

jan.…

feb.…

mar…

apr.…

mai…

jun.…

jul.12 aug…

sep.…

okt.…

nov…

des.…

jan.…

feb.…

External meaningless tags, share of total external tagseach month

External empty tags, share of total external tags each month

Figure 19 Empty and meaninless tags applied by external users, in percent for each month

1. Words that belong to a lay vocabulary:

Words concerning cancer or medicine in general that are Danish and easily understood for the general Danish public belong to this category. Compound words that are combinations of Danish and Latin/English were usually put here.

Not only because they are lay compound words in their origin, but also because the Cancer.dk editors stated that they could not use these combinations in their texts. They regard them as wrong, or at least not optimal. These words may also be used by professionals, but there are often one or a few professional synonoms belonging in category 3.

Examples: Kræft (cancer), Brystkræft (breast cancer), Brystcancer (breast cancer [a combination of Danish and Latin/English]), Mammakræft (breastcancer [a danification of the latin Cancer Mammae, or a way to describe breastcancer]).

2. Words that are neutral and common to both laypersons and professionals:

Words covering concepts that may be used in connection with cancer, but also words related to other topics belong to this category. These are often concepts where professionals and laypersons generally use the same concepts. The category includes common names for symptoms and parts of the human body. It also includes names of fundraising events.

Examples: Rygning (smoking), Arvelighed (inheritability), Medisinsk behandling (medical treatment), Transport (transportation), Strålebehandling (ray therapy), Kalk (calcium), Årsager (causes), Pårørende (next of kin), 50 år (50 years), Stress (stress), Appelsinkage (orange cake), Armbånd (bracelet), Bryst (breast), Diætist (dietitian), Klinisk forsøg (clinicial trials), København (Copenhagen), Pinkcup [name of a combined women’s golf tournament and fund raising campaign, in Denmark], Mindfullness (mindfulness).

3. Words that belong to a professional’s vocabulary:

This category includes Latin or English words concerning cancer or medicine in general. Danish words concerning cancer or medicine in general that are not well known by the general Danish public are also included.

Examples: Herceptin [a medicine], Breast cancer, cancer mammae [Latin for breast cancer], Antihormoner (antihormones), Statiner [a medicine], Triple cancer [short for Triple-negative breast cancer], Bløddelssarkom (soft tissue sarcoma).

4. Words that are hard to put in the previous three categories:

This category is used for names, tags related to membership in the Danish cancer society and the use of forms at Cancer.dk. Many of them have a meaning not related to medicine at all.

Examples: afmelding nyhedsbrev (newsletter unsubscription), adresseændring (change of address), alt afmeldes (unsubscribe everything), anja olsen [a woman’s name], arrivederci franz [the name of a music album], sommerlotteri (summer lottery).

5. Words that were not categorized at all:

Tags that do not contain real words, empty tags and tags with no meaning are put in this category. Tags that looks like usernames are also in this category.

When categorizing, it was sometimes hard to decide which category to choose. With names in category 4 and usernames in category 5, it was necessary to decide wether a halfway-finished name was a username or a name. I ended up putting e-mail addresses in category 4, because they usually identify a person in the same way as a name does.

And usernames were kept in category 5 because there were so many tags where it was hard to say whether it was a username or something meaningless. However, in the facet categorization, which Bing and I did after the lay/professional categorization, we did try to separate between usernames and meaningless words. Thus, we covered this distincion and made it possible to discover if the taggers turned out to have a separate behaviour concerning this.

It was also sometimes hard to decide wether some tags chould be in category 2 or 4.

One example is the fundraising projects. I see them as neutral: they are related to cancer but are generally not a part of the cancer medicine field. On the other hand, I have put words concerning membership in the Danish Cancer Society in category 4.

This is because I see these words as more peripheral to the field of cancer. But the distinctions leave some terms to fall between two stools, like trækningsliste [list of winning numbers in a lottery] or engangsbeløb [lump sum]. Usually, we put these tags in category 2. In general, we restricted category four and five to the kinds of tags mentioned in the description and used category two when in doubt.

Number of tags Share Number of tags beforefeature change Share of tags beforefeature change Tags on day of featurechange Share of tags on day of feature change Number of tags afterfeature change Share of tags after feature change

1 Lay 3322 13,15 3253 13,20 8 21,62 61 10,59 2 Neutral 11777 46,64 11601 47,08 21 56,76 155 26,91 3

Professional 1988 7,87

1955 7,93 0 0,00 33 5,73 4 Hard to

categoreize

2075 8,22 1968 7,99 1 2,70 106 18,40 5 Empty tags,

meaningless tags, etc.

6091 24,12 5863 23,79 7 18,92 221 38,37

25253 100 24640 100 37 100 576 100 Table 7 Number of tags in lay/professional categories in total and before feature change, on day of feature change and after feature change

Table 7 shows the number of tags in each of the five categories, and the percentage share of each category. The first colomns with numbers give total numbers, then numbers are given for whether the tag was applied before the feature change, on the day of feature change, or after the feature change. Numbers for the day of the feature change are small, but the status of the feature for these tags is unknown. Thus, they are separated from the other tags.

All the tags, whether they belonged to a lay or a professional vocabulary, constituted the total amount of tags available on Cancer.dk. Thus, this was what users could see.

But with internal and external tags separated, one can see who applied which tags.

Table 8 show how internal and external users applied words from a lay and professional vocabulary as tags. The table include both numbers and shares. 37 tags from the day of the feature change and 1 tag from an internal tagger after the feature change are excluded. The most visible differences in tagging pattern is between internal and external taggers. But there are also differences in tagging behaviour for external taggers before and after the feature change.

Tags before feature change Tags after

Table 8 Tags for internal and external users with lay and professional vocabulary, before and after feature change. 1 tag from an internal user after the feature change and 37 tags from the day of the feature change are excluded.

Figure 20 shows the numbers of lay, neutral and professional tags without categories 4 and 5, empty and meaningless tags and tags that are hard to categorize, from Table 8. For external taggers, the most visible change is a higher share of lay tags and lower share of neutral tags after the feature change. For internal tags, the share of neutral tags is higer than external tags. And internal tags applied before the feature change have relatively more lay tags than professional tags, compared to external tags applied both before and after the feature change.

Figure 20 Share of lay and professjonal tags from internal and external users before and after feature change. 38 tags are excluded.

Both groups applied a high share of neutral terms. Both groups also applied more lay terms than professional ones. I assume that most of the external taggers were laypersons when it came to medicine in general and particularly cancer medicine. But the external taggers applied the highest share of the professional terms Before the feature change, 6.5% of tags from internal taggers were professional, and 12.4% from external taggers, a share almost twice as high.

After the feature change, external taggers applied fewer neutral tags and more lay tags.