To elicit information on employee attitudes, we used a questionnaire that we distributed digitally. The questionnaire consisted of 56 items with open and closed questions. Questions were compiled based on background information (for instance, questions mapping the employees’ languages spoken and how long they have lived in Norway), to determine their awareness of current language policies at the institution (e.g., if they are aware of existing policies), and open questions to express their needs and opinions. The online questionnaire was distributed on the university’s intranet to all employees and students. It was available in English, Bokmål, and Nynorsk (the two official written variants of Norwegian). The study was approved by Sikt, the Norwegian Agency for Shared Services in Education and Research. The participants were fully anonymized, and the data were stored on a secure server that only the researchers in the project had access to.
We received 705 responses from employees, which accounts for 9% of the total employee population (n = 8,051). Seventy percent of the respondents held a teaching/research position (e.g., Professor, PhD, Postdoc), and the latter 30% consisted of administration, maintenance, HR, IT, and technicians. Approximately 56% of the participants were born in Norway, and 43% outside of Norway. Fifty-five different native languages were reported by participants.
The data were analyzed quantitatively (descriptive statistics) in closed questions, and qualitatively (content analysis) in open questions. In the current study, 7 questions were used in the analysis pertaining to attitudes toward language policy. We first asked about awareness of language policy, then asked questions about how NTNU should practice language policy guidelines. For the quantitative analysis, we considered the responses toward seven statements relating to language policy which asked people to indicate agreement on a 5-point Likert scale. The responses of all three questionnaire versions were matched and analyzed descriptively via relying on absolute frequencies.
For the qualitative analysis, we investigated three open questions (what participants thought about the language policy, if there were anything they would change about the language policy, and if they had any additional comments). We received responses from 219 participants. The data were analyzed in MaxQDA through qualitative content analysis. The two raters created open codes, then merged these codes into larger themes and double-coded for reliability using the so-called “Gioia method” (Gioia, Corley, & Hamilton, 2013).