JCPSLP Vol 18 no 2 July 2016

Table 5: Presence of recency statements and citations for each type of information site

Purpose

Recency

Citations present

None

>5 yrs

<5yrs

Total Stated

N

Y

Guidance/resources

20

3

14

17 (46%)

32

5 (14%)

Product/technique overview

50

7

31

38 (43%)

64

24 (27%)

Opinions and news

1

2

7

9 (90%)

2

8 (80 %)

Total

71

12

52

64 (47%)

98

37 (27%)

Table 6. Domain types for each category of information site

Purpose

Non-commercial

Total

Commercial

Other

Total

N/S

.edu

.org

.gov

.com/.co

.net

Guidance/resources

1

12

3

16 (43%)

19

1

1

21 (57%)

0

Product/technique overview

4

37

3

44 (51%)

24

7

11

42 (48%)

2

Opinions and news

0

3

0

3 (30%)

6

1

0

7 (70%)

0

Total

5

52

6

63 (47%)

49

9

12

70 (52%)

2

Credibility and accuracy are not the same thing The credibility of a website becomes particularly salient to those searching unfamiliar topics (Eastin, 2001). The indicators most strongly associated with seekers rejecting a website include a lack of authorship or source information, a heavy commercial focus, and the presence of information the seeker knew to be incorrect (Rice, 2006). Information websites evaluated in the current study showed high variability in their external credibility, with sites from several reputable organisations lacking both citations and recency statements. Several researchers have pointed out the inherent limitation of putative credibility cues such as source and author attributions, which can increase apparent credibility but are no guarantee of quality (Eastin, 2001; Kunst & Khan, 2002). Other credibility cues such as levels of evidence and the presence of referencing present similar risks given that, without peer review, evidence and citations may be falsified or cherry-picked to support inaccurate information. Furthermore, common indicators of credibility have been shown at best to have only a weak correlation with each other (Kunst & Khan, 2002). This finding was mirrored in the chi-squared analyses from the current study for the credibility indicators of recency and references, suggesting that single measures may be insufficient predictors of a website’s credibility or accuracy. In light of these findings, training AAC consumers and their families to “evaluate” on-line sources based on their apparent credibility alone may not guarantee the retrieval of reliable information. Given that approximately 35% of information seekers do not consult with their health professional about the information they find on-line (Fox & Duggan, 2013), it is important for AAC professionals with expert knowledge to initiate a dialogue about information seeking with health consumers and their families (Roche & Skinner, 2009; Zaidman-Zait & Jamieson, 2007). At a broader level, official information hubs or wikis, such as the proposed Newell network or the NDIS hub (http://www.ndis.gov.au), could be used to disseminate evidenced-based information on communication disability, intervention techniques and available products, in addition

Judging by our findings, it is likely that naïve searchers will encounter limited information about AAC simply through the use of generic terms such as communication app , but may be less likely to find this information when diagnosis-specific search terms are used. Even when relevant to AAC in general, the specific focus of many websites returned through Google may be a poor match for those in search of practical advice. For instance, only 4% of results from device-specific searches addressed issues of implementation. Consequently, searching for a specific AAC solution may unearth a plethora of information on where to purchase it, but little on its suitability, viability, or optimal use. Many of the products and services advertised on-line were also inaccessible to Australian searchers. Availability of such information is critical to informed decision-making and implementation success, particularly under the new NDIS. In anticipation of this problem professionals should direct consumers to appropriate information sources (e.g., specific websites or information hubs) as necessary. As on-line knowledge brokers, AAC service providers and support organisations can play an important role in supporting the NDIS’s ILC domain by optimising the reach of quality on-line content. Particular attention should be paid to page titles and salient keywords, which can improve chances of search-engine retrieval. Internet- based health information is frequently inaccessible to people with disability and to those from non-English speaking backgrounds, and often requires high levels of reading proficiency (Greenberg, D’Andrea, & Lorence, 2004; Zaidman-Zait & Jamieson, 2007). Given that accessibility issues will affect a large number of NDIS service users, attention to these features in website design is of paramount importance. Guidelines on web content accessibility can be found at www.w3.org/WAI/intro/wcag. php . By attending to issues of accessibility and visibility, reputable knowledge brokers can help to improve access to local and high-quality information for Australian AAC consumers on-line.

72

JCPSLP Volume 18, Number 2 2016

Journal of Clinical Practice in Speech-Language Pathology