Data Collection

1. Recruitment

We conducted surveys among organisations in Australia and China. In phase I (early 2016), survey companies were hired to conduct the data collection. Because this study examines the status of IP adoption within organisations, participants would only qualify for the survey if they passed a number of screening questions. In total we received 288 completes (Australia 138; China 150). In phase II (mid-2016), an invitation to the survey was sent to APNIC members, the AUSNOG mailing list and the CNNIC mailing list; 98 participants (Australia 60; China 38) completed the survey. Since in phase II we only advertised the survey to people that are knowledgeable about IPv6, no screening questions were asked.

We used two different ways to recruit participants with the aim to get a more diverse sample. Some previous studies only advertised their surveys via the RIRs, but then they were usually answered only by people from the telecommunications industry or large organisations. Using survey companies necessitates the use of robust screening question to filter out potential participants that lack the required knowledge.

2. Data cleaning

The similarities of responses for each country were then checked to detect potential data records from same organisations. Two responses may come from the same organisation if they 1) have the same domain name (voluntary question the survey), 2) came from the same IP address (as logged by the survey web site), or 3) have quite similar answers regarding the demographics of the organisations and deployment of IPv6. After identifying similar data records, the authors then discussed and decided whether to treat them as duplicate records for one organisation. When duplicates were removed, the records removed were picked randomly. We also removed a few records where respondents stated that IPv6 was only used in a test environment but then in a later question stated that the volume of IPv6 traffic is equal or above the volume of IPv4 traffic.Nearly all of these are respondents from market research company (only one exception for China). With contradictory answers, our trust in the records was low and we decided to remove them.

3. Demographics

The total number of participants in both countries recruited by both approaches was 292 (Australia 157; China 135). Around 75% of our participants were CIOs, top-level managers or mid-level managers (Australia 77%; China 75%), while 25% were low-level managers and IT administrators/technicians. The vast majority of them (Australia 93%; China 91%) said their decision making authority with regards to networking infrastructure was final or they have significant decision making influence. The senior level and decision making authority of the respondents means that they should have very good knowledge of the status of IPv6 in their organisations. Consistent with the senior positions, most of the respondents fall in the age bracket of 30--50 years.

Figure 1 shows the number of organisations by ANZSIC industry type. The majority of our organisations are from the media and telecommunications industry. This is expected because (1) APNIC members are biased towards this industry and (2) the screening question used for the market research companies required participants to know about IPv6, which also creates a bias towards this industry. Nevertheless, only one third of Australian organisations and slightly over 40% of Chinese organisations fall into this category. The rest comes from a variety of different industries, with the other larger categories being Manufacturing, Retail (Australia), Professional, Scientific and Technical Services, Education and Training and Healthcare (Australia).

EXIT

Of the “Media and Telecommunications” organisations (MTOs), the majority were Telcos or ISPs (Australia 71%; China 85%), while 17% (Australia) and 8% (China) classified themselves as “Other Provider or Reseller of IP Services (e.g. Research/Academic Network)”, and the remaining organisations classified themselves as “Other”. Most of the MTOs offer fixed-line and mobile services (Australia 57%; China 75%), while a significant number only offer fixed line services (Australia 35%; China 25%). The remaining providers offer only wireless services.