Running Your Study in Multiple Languages
If you are sending your survey to respondents who speak different languages (for example, English and French speakers in Canada), and need to have the appropriate version displayed to different respondents. Conjointly has automatic translation functionality that can help you pre-translate the survey before professional translators are engaged.
Setting up automatic translations for your study
Here are several simple steps to set up the automatic translations:
Find translation settings under Advanced settings and click on to launch a pop-up:
Click on “Add language and automatically pre-translate as much as possible” to create a new language for your experiment using pre-existing translation from a library.
Select the language that you would like to base your translations on. In the example below, we have set “English (USA)” as the base language. Select the new language (here we are going to create German language) and clickwhen you are ready.
Doing so will create two new functions in your experiment:
Language filter: You can now toggle the language that you would like to view and edit the text within the same interface. To illustrate, in the experiment below:
Translation tab: This interface lists all text in the survey, including conjoint inputs, additional questions and options, default text (i.e. introduction, hints, etc.) as well as buttons.
NB: If you add new questions to your survey, remember to update your translations.
Other notes on automatic translation
You will be prompted to top up your prepaid balance if it is insufficient to cover the translation cost.
With the automatically pre-translated text from step 2 above, all default text and buttons have been professionally pre-translated for various countries: , , , , , , , , , , , , , , , , , , , , , , .
With the automatically pre-translated text from step 2 above, newly added text (conjoint attributes and levels and additional questions) are automatically translated (not by humans). Please review these items carefully.
You can preview the survey across various languages by clicking on the preview buttons (highlighted below).
Hyphenation checkbox: When this checkbox is ticked, the survey text will be automatically adjusted to fit the text into the area/page, even if it needs to break some words and hyphenate them. If you prefer to present whole words, you can untick this checkbox.
Right-to-left (RTL) text direction: This checkbox should be ticked when the language is written from right to left (e.g. Arabic).
Default experiment language: Default experiment language affects the language used in the reporting interface. If you want the report to be displayed in English, then the default language needs to be set to English.
How language detection works
There are two options for how language is selected for respondents:
Automatic detection using browser settings. In this case the Conjointly survey will look at the respondent’s browser locale setting and will direct them to the right language.
Manual choice of language to the respondent. In this case, each respondent will need to select the right language for themselves (see example).
However, both settings will be overridden if the link contains the GET variable “locale”. For example, to display a survey in Russian, you can add ?locale=ru to the survey link (e.g. https://run.conjoint.ly/study/8356/ts1kg8ri1i?locale=ru). To display it in English, add ?locale=en.
Should you require professional translation of your survey into other languages, please do not hesitate to reach out. We can organise a speedy high-quality translation. Approximate timeframe is 1 to 3 business days (depending on complexity). Translations typically cost from $200 per language per study.
You can also export the survey in XML format and send to your own preferred translation agency.