Any translation company can tell you that translating survey responses is a common occurrence and that budgetary constraints are often a consideration when undertaking this type of work. Because most survey projects have thousands of responses in many languages, the highest level of service (translator and independent editor) may not fit within the customer's budget. However, that does not mean that there is no solution that will meet the needs of the customer.
The most important thing for a buyer to consider when translating survey responses is the specific outcome you are trying to achieve. Are these responses going to be published? Are they going to only be viewed internally? Are they strictly for research/knowledge purposes? Regardless, it is important to consider and communicate what you are looking for with your translation provider. They can then match the level of service that fits both your needs and budget.
Translating survey responses brings to light a very important topic in the translation industry – levels of service. Read on for the four most common levels of service you can expect to see.
Typically the highest or “best” level of service that a provider will offer. It utilizes a human translator and human editor, both with similar qualifications. There are minimum requirements that a translator must possess under the ISO 17100 standard (which Argo carries).
This is a slightly lower level of service, but still a viable option. It would rank as “good” in terms of accuracy. Rather than having two linguists working on the survey responses, the translator will edit their own work. This is a good option for budget-conscious customers, as it will still outperform any machine translation.
This level of service would be categorized as “good enough”. In essence, the survey responses would be translated by an automated machine translation engine (such as Google Translate) and then reviewed by a single linguist. As outlined above, there are better options, and typically the cost of machine translation with post-editing is on par with translation with self-editing. This should be avoided if at all possible.
We’ve finally reached rock bottom. As the translation industry has observed time and time again, machine translation (MT) simply cannot match a human translator. MT lacks the subtlety of language and is typically not consistent. Another drawback of this level of service is that it does not account for context which is always key for an accurate translation.
This level of service is typically appropriate for research work or if the potential survey responses are simple in nature and of low value. This is also appropriate if you have a huge number of responses. A common approach is to use MT to quickly and inexpensively translate the content and then sort through the material. If any of the responses need clarification, a human translator can adjust those responses.
It is also important to note that, as the list of levels of service descends, typically the price drops as well. It is incredibly important to know what level of service you are receiving from your provider, as it is directly correlated to the price of translating your survey responses. This becomes especially important when comparing providers.
Another very important point to consider when buying translation is transparency. When you as a buyer review the translation quote or proposal from the supplier, do you know exactly what you are getting for your money? Does it list what services you will be receiving, and what those services entail? Knowing what you are getting, and perhaps more importantly what you are not getting, is incredibly important, especially in regard to translation services.