Talk:Fundraising 2009/Survey/Translations

From Meta, a Wikimedia project coordination wiki
Please add questions below.

Education question[edit]

2.5 The highest level of education I have achieved is: grade school or less / some high school / high school graduate / some college / trade, technical, or vocational training / college graduate / some postgraduate work / post graduate degree

We ran into some problems with this on nl/fr, this doesn't really work between most countries (even ones that share the same language). The systems are just too different, here are some solutions we came up with (these are excerpts from a large conversation, they're not in order):

  • <effeietsanders> so then I think it is better to get rid of the whole question
  • <effeietsanders> or make it US only
  • <effeietsanders> anyway, you could either go much more detailed
  • <effeietsanders> or much more zoomed out
  • * kibble would prefer much more zoomed out
  • <effeietsanders> like... primary school, secondary school, college/university, other

Any comments on these solutions? Cbrown1023 talk 21:19, 25 July 2009 (UTC)[reply]

As an addition, all seem acceptable, although "us-only" and "leave out" would mean that we'd end up having data without educational-level. So zoom-out would be the best option, just to make some more general levels. Although it would be very difficult to get educational levels that mean about the same in all countries (for comparison) and aren't too broad. Best regards, m:Mark W (Mwpnl) ¦ talk 21:23, 25 July 2009 (UTC)[reply]
I also ran into the same problem... Zoom-out would be better. I wonder if that's enough (and mostly fitted) with high school or under/college/graduate school or so.--Takot 07:17, 26 July 2009 (UTC)[reply]
This is a fairly complex issue. I did a little research what is generally used to measure education levels on international surveys and found there are basically two scales: CASMIN and ISCED97. The ISCED97 was developed by UNESCO and seems to be used more frequently ISCED97. The scale is more detailed than we need but perhaps we can collapse some of the categories. I will propose and alternative here shortly. jpilisuk 20:37, 27 July 2009 (UTC)[reply]
Proposed Revision - The highest level of education I have achieved is: Some Primary or Secondary education (full-time compulsory education, not completed) / Secondary education (high school, full-time compulsory education, completed) / some Tertiary education (college, university, not completed) / Post-secondary, non-tertiary education (trade, technical, or vocational training) / Tertiary education - first stage (college, university, completed degree) / Some second stage tertiary work (advanced research degree, Masters, PhD, not completed) / Tertiary education - second stage (advanced research degree, Masters, PhD, completed) I'm not sure this works very well. I may need to make the answers more brief and move the definitions of "Secondary," "Tertiary," etc... to the end. Any other ideas? jpilisuk 23:01, 27 July 2009 (UTC)[reply]
Thanks Jeff, this sounds like something that could actually work! This gives the opportunity to translate it to the Dutch equivalents, without losing value when comparing the results. If this is okay for the others, I'll translate this Mondayevening. m:Mark W (Mwpnl) ¦ talk 18:05, 2 August 2009 (UTC)[reply]

marital and age question[edit]

I'm working on Fundraising 2009/Survey/Translations/ja. I'm wondering if we could choose "Prefer not to say" about age or marital status... IMHO, those are kinds of sensitive ones, even we don't answer with our User IDs or names. --Takot 05:31, 26 July 2009 (UTC)[reply]

As far as I know, all survey responses will be anonymous so this should be a problem. Demographic information like that is pretty important, so it would be nice to include those questions without the "prefer not to say" answer. (Remember, the survey is going out to past donors and future donors, so most Wikimedians wouldn't be the ones who would fill out the survey.) Cbrown1023 talk 13:57, 26 July 2009 (UTC)[reply]
Ah, the same question came into mind here, but then concerning the "household income" question. I'm not sure how this is in other cultures, but in the Netherlands marketing experts are faced with the problem that Dutchies do not like answering questions about their money - not even if it's anonymous. In fact, studies showed that quite some people stop the questionnaire if they can't answer "Prefer not to say". I understand the need to have as much demographic information as possible, but if it resuts in less completed forms that might be counterproductive. I can imagine this goes for all personal questions (including the one about the marital status). Best regards, m:Mark W (Mwpnl) ¦ talk 14:05, 26 July 2009 (UTC)[reply]
Ideally we want to get answers to all these questions. However, if there is evidence that a significant number of individuals may abandon the survey unless there is a "prefer not to say" option, then we should add it. I'd like to address this issue. Which questions really justify this change? Marriage status? Household income? Age? Education? Employment status? jpilisuk 21:31, 27 July 2009 (UTC)[reply]
I think it would be best to have a "prefer not say" option for all personal questions, but the most important questions are (I'd say) marriage status and income. Although - just to be sure - I think it's best to give the "not say" option too all the other personal questions as well. m:Mark W (Mwpnl) ¦ talk 18:02, 2 August 2009 (UTC)[reply]

currency conversion[edit]

What if we translate currency converted in advance and propose it in local currency, rather than preparing a link to currency converter and keeping it USD? I understand that currency exchange rate varies day by day, but not so drastically over one night. Moreover, we are asking about past (18 months) donation. Answers, I suppose, will be counted as computational way, such as branch-ID or something, so you don't have to convert it again.

From other viewpoint, it sounds good where (specific) language users and residential country (mostly) match one-to-one, such as Japanese-Japan, but will not work among other widespread languages over countries or lingua franca. --Takot 05:44, 26 July 2009 (UTC)[reply]

From my point of view, that makes sense. We could even put both there (ie. USD/JPY), just in case we have someone like a Japanese living in San Francisco (:-)). Cbrown1023 talk 14:00, 26 July 2009 (UTC)[reply]
For specific languages where there is a one-to-one match with a country (eg, Japanese-Japan), I agree this makes sense. Thanks for the suggestion. jpilisuk 20:37, 27 July 2009 (UTC)[reply]