Report 1 on the action Translating and testing a victimisation survey module
by
National Council for Crime Prevention, Sweden
1 Introduction
This report from the Swedish National Council for Crime Prevention (Brottsförebyggande rådet) is being written in accordance with Draft Grant Agreement No. 11002.2008.003-2008.211, and relates to the action Translating and testing a victimisation survey module. It is the first of three technical implementation reports to be submitted by the Council in connection with the action, as outlined in Annex IV of the Draft Agreement signed and returned to Eurostat on September 15th 2008.
The report relates to the activity of translating the victimisation survey module from English to Swedish and the translated full text of the victimisation survey module is included as annex.
1.1 Organisation of the report
The report is organised in the following way. The following section briefly describes the background to the National Council's use of the translation agency subcontracted to translate the questionnaire into Swedish and the instruction given to the translation agency in connection with the contract. It then describes the subsequent work conducted by the National Council itself to review and adjust the translation where this has been deemed necessary.
The report then proceeds to provide a description of the issues arising in connection with the translation of the questionnaire into Swedish. Each section of the questionnaire is examined in turn, and the issues raised by the subcontracted translator are presented along with those that emerged in the National Council's subsequent review of the translation.
1.2 Subcontracting of the translation and its subsequent review by the National Council
Swedish public sector tendering regulations are such that the National Council only has rather limited options with regard to the choice of translation agency employed in connection with work of this kind. It is not the National Council itself that conducts the tendering process, and we have yet to find a translation agency among the possible alternatives that has shown itself to be entirely satisfactory for the agency's needs. However, at the same time this means that we are used to receiving translations that almost always require careful checking and further work before we consider them satisfactory. Furthermore, the Research and Development Unit at the National Council has native speakers of both English and Swedish with several years experience working in a variety of areas of criminological research, including both national and international victim surveys.
The work conducting the initial translation of the questionnaire from English to Swedish was subcontracted on 24 September 2008. The instruction to the translator stated amongst other things that the Council required the translator to make a careful note of any difficulties experienced in the course of the translation, particularly in cases where the English appeared ambiguous, or where it was not clear what constituted the correct or best translation into Swedish. In such cases the translator was also asked to suggest possible alternative translations of terms or concepts regarded as problematic. The National Council received the translated questionnaire on November 10th and a page of queries and notes from the translator the following day. This initial translation of the questionnaire was assessed by the National Council to be of reasonable quality, although not exceptionally good throughout.
The translated questionnaire has subsequently been reviewed item by item by two research analysts working at the Research and Development Unit of the National Council, the one a native English speaker with a PhD in criminology and 10 years experience of his own translating criminological texts from Swedish to English and vice versa, and the other a native Swedish speaker, also with a PhD in Criminology, and with 12 years research experience working with international comparisons of crime and victimisation trends, and with large-scale national victim surveys in both Sweden and South East Asia.
The points presented and discussed below in connection with the various sections of the questionnaire are based on both the notes received from the translator and the subsequent review of the translation conducted by the National Council. Minor adjustments made to the initial translation – such as the final choice of a specific term where two or more reasonably similarly adequate alternatives were available are not discussed.
2 Issues emerging in connection with the translation and its subsequent review by the National Council
2.1 Section A
Question A7 - “married or living in a registered partnership”. In Sweden, there are a very large number of unmarried couples are who are “samboende” (cohabiting) - which for those concerned has a similar status to being a “registered partnership” (although the legislation relating to e.g. who inherits what in the case of death etc. has not yet caught up). Such relationships do not constitute a “registered partnership” in the strictly legal sense, but both Swedish “couples” and Swedish questionnaires for the most part regard this type of cohabitation as more or less corresponding to “marriage”. Questionnaires therefore usually include “cohabiting” it in the same response alternative as “marriage” – e.g. “married/cohabiting”. We have a slight concern that this may cause confusion in connection with the current questionnaire, since no mention is made of cohabiting until the next item in the questionnaire.
Question A12 – response alternative 7: “permanently disabled”
We were unsure here whether to use a strict translation of “permanently disabled” or whether to use the official labour market terminology, since all persons who are permanently disabled and unable to work as a result are referred to as either “Förtidspensionerad eller avtalspensionerad”. Since it is the latter term that is usually used in this context in similar Swedish surveys, this is the option we have chosen in the current context.
2.2 Section B
Question B3 – “How safe do you feel walking alone in the area where you live....?”
The response alternatives are rather “asymmetrical” - you can either feel “very safe” or “quite safe” - but if you feel unsafe, then it's either “a bit unsafe” or “very unsafe” - we have translated “a bit unsafe” faithfully into Swedish, and this is the formulation that will be tested by the subcontractor who will be conducting the cognitive testing etc. – but we wondered a little why this formulation had been chosen to begin with. The National Council's own national survey on levels of safety and worries about crime employs the response alternatives: very safe, quite safe, quite unsafe, very unsafe.
Question B6 – Ditto
Questions B11 and B12 – “worried about having your car stolen or damaged or something stolen from it” - Here the initial translation of “damaged”, which faithfully followed the English original, sounded as though it could be misinterpreted to include damage caused e.g. in an accident on the road or when parking. We have therefore used a formulation that corresponds to the English: “subjected to vandalism” - this is a formulation we have successfully used in our own Swedish national crime (victim) survey.
Questions B12 and B13 – the questions refer to worries about “being physically attacked by strangers”. The translation here was a little tricky – several different translations of “strangers” are possible – the initial translation “främmande människor” captures much of the sense of “strange” associated with the word “stranger” - but it can also be used to refer to people with whom one is acquainted. We've finally decided to instead opt for a term that we feel better captures the sense that the question is asking about people the individual in question is not acquainted with. The best translation of the term employed back to English would probably be “people that you don't know”. (In the final section of the questionnaire, where questions are asked about violence at the hands of a stranger, we have felt comfortable about using the translation “främmande människa” because the questionnaire at that point includes a definition for the respondent of what the term is intended to mean).
2.3 Section C
No special comments
2.4 Section D
Question R1 - “How many offenders were there” - The question is very simple, but we have chosen to use: how many offenders “participated”, simply because it works and sounds better in Swedish without altering the content or intent of the question in any significant way.
Question R2 – In the English version we would suggest that “except the offender” should be replaced by “besides the offender/offenders”. In the Swedish version, since the previous question indicates the possibility of more than one offender, we have translated as though the original did in fact state “offender/offenders”.
Question R7 – The English version suddenly assumes there was only one offender again - “do you think the offender was under the influence of alcohol or drugs” - would it not be better to ask if “any of the offenders” were under the influence of alcohol or drugs? We have translated faithfully and not made any adjustment to the Swedish version in this instance.
Question R11 – Response alternatives “attempted to rape” and “sexually assaulted” are very difficult to differentiate – both for the respondent and we would imagine when it comes to working with the data. We have translated “sexually assaulted” is if the original was “other form of sexual assault”
Question D20 – Response alternative “belongs to my job” - here we were forced to expand the phrase in the Swedish translation so that it became “the object/objects belong to my job” - a faithful translation of “belongs to my job” would have meant using a phrase that in Swedish also means “it's part of the job”.
Question D23 - “Did the police give you a written document or a crime number of the incident” - Here we have altered the question somewhat to correspond to police practice in Sweden and asked whether the respondent received a “written copy of the offence report” or an “offence report number” from the police.
Question D25 and D32 etc. – We wondered once again why respondents can be “Fairly or Very satisfied” but only “A bit or Very dissatisfied”
Question D30 and D31 etc. – Response option 2 “Practical help (e.g. completion of insurance or criminal liability claim forms)” – We're not sure what a “criminal liability claim form” is. Having first translated the term as best we could, we finally decided to remove it completely, feeling that it was unlikely to mean very much to Swedish respondents – and that it was thus likely only to cause confusion.
2.5 Section E
Question E23 – Noted a shift here from previously always asking “Did this happen in 2007” to asking “Did this happen last year” - (we have translated as “last year”) – the next question, E24, then reverts to “How many times did this happen in 2007”.
Question E26 – response alternative 1 “Did not receive at all the service” – our translator found this a bit difficult – unusual formulation in English – not sure whether it meant “did not receive the service at all” or was a typo for “did not receive all the service (i.e. “the service in its entirety”). In the final translation we've guessed it meant “Did not receive the service at all”.
Question E30 – slight concern about the use of “in your country” - not entirely clear that this formulation will necessarily be perceived in the same way by individuals who've lived in the same country all their lives and e.g. recent immigrants.
Question E50 - “national insurance number” - not used in Sweden. Here we have translated as “personal identity number” - which is unique to every individual and is used in much the same way.
Question E54 - “...how did the misuser get to know your personal details” - We have not used “misuser”, since when translated into Swedish the equivalent term is synonymous with “drug user/abuser/addict” - we have instead used the Swedish term for “fraudster/deceiver/person who commits fraud”
2.6 Section F
No special comments
2.7 Section G
Q1 and Q2 – response alternative 8 - “Followed or stalked you” - Q1, response alternative 8 does not include “so that it frightened you” – Q2, response alternative 8 does include “so that it frightened you”. We wondered why this difference? (Have translated faithfully so the response alternatives are different in the translated version also)
Q7 – response alternatives 2 and 3 – miss individuals aged 45 - “Between 25 – 44 years”/ “Older than 45 years” - individuals aged 45 don't fit into either of these categories – we have translated response alternative 3 as “Aged 45 or over”
Q 8 – There are two response alternatives numbered “10” - “concentration difficulties” and “annoyance”
Q20 – See Q7 – same thing here.
Q23 – see D23 – same thing here
Q62 and 63 – see D30 and 31 – same thing here
Q88 – English formulation very strange – but we think we've been able to guess what was being asked.
Q95 – see Q7 – same thing here
Q113 – response alternatives appear a bit strange once again – there is a fairly large gap between “at least once a month” and “once or a couple of times a year”.
2.8 General points
In questions asking about age of e.g. offenders, it may be useful to include an instruction to the respondent that if they're not sure, then their best guess will suffice.
Another more general point is that, prior to Section G, there do not appear to be very many response options indicating that the respondent may either not know the answer to certain questions or may choose to decline to answer certain questions. Is this intentional?