2.2.2. Newspapers

To answer the research questions, the University of Calgary's Proquest online database *Canadian Newsstream* was used as a source, a database consisting of *n* = 300 English Language Canadian newspapers, for its complete time range from 1980 to March 2018. Canadian newspapers were chosen because (a) existing and potential speech-related professionals are readers of newspapers, (b) the database comprises over 300 news sources including all regions of Canada, (c) over 75% of Canadians still read newspapers [52,53], and as such are influenced by what they read and (d) parents, teachers, and career counselors who often give advice on career ideas to young adults also are readers of newspapers [54].

To maintain a clear and feasible scope [48], eligible newspaper articles were identified using explicit search strategies [50]. On March 28th, 2018 four search strategies were employed.

Strategy 1, The newspaper database was searched for the terms "speech language" OR "speech patholog\*" OR "speech therap\*" OR "audiolog\*" in combination with "neurosc\*" OR" "neurotechn\*" OR "neuroen\*" or "neuro\*" obtaining *n* = 403 unique results articles,

Strategy 2, The newspaper database was searched for the terms "speech language" OR "speech patholog\*" OR "speech therap\*" OR "audiolog\*" with 49 neuro-related terms (Table 3) omitting cochlear implant obtaining *n* = 83 unique results,

Strategy 3, The newspaper database was searched for the term "speech technolog\*" obtaining *n* = 438 unique results, and

Strategy 4, The newspaper database was searched for the terms "speech language" OR "speech patholog\*" OR "speech therap\*" OR "audiolog\*" with the term "cochlear implant\*" obtaining *n* = 321 unique results.

The newspaper articles found for each of the four search strategies were downloaded as individual PDFs and imported into Atlas.TI8™, a qualitative data analysis software application for qualitative, thematic content analysis.

## *2.3. Data Analysis*

## 2.3.1. Newspapers and Academic Newspapers Together

A descriptive quantitative and thematic qualitative content approach was employed to answer the research questions using the software's ATLAS.Ti8™ Adobe Acrobat. For the descriptive quantitative data analysis, hit counts were generated for the presence of terms representing various speech-related fields and professionals in relation to neuro-related terms. For the qualitative content analysis both authors read the content of academic abstracts and newspaper articles. While reading the content, both authors generated codes using a term or phrase reflecting a given role seen as evident in a given content. The coding was deductive in the sense that the top-level theme examined (role) was predefined by the scope of the study [55,56]. The actual roles found, however, that emerged from reading the academic abstracts and the full-text of the newspaper articles were not pre-set and as such could be seen as an inductive approach [55,56].

## 2.3.2. Trustworthiness Measures: Newspapers and Academic Literature

Trustworthiness measures include confirmability, credibility, dependability, and transferability [57–59]. Differences in codes and theme suggestions of the qualitative data were few, discussed between the authors, and revised as needed to ensure credibility and dependability. Confirmability is also evident in the audit trail made possible by using the Memo and coding functions within ATLAS.Ti8™ and the sticky note function in Adobe Acrobat. As for transferability [57–59], the methods description gives all required information for others to decide whether they want to apply these keyword searches on other data sources such as other newspapers, or social media such as twitter or other academic databases or want to expand on the keywords used.
