Examining the User Acceptance of Gigabit Broadband Service: The Case of UC2B
Posted: 1 Apr 2015
Date Written: March 31, 2015
The focus of this research paper is to examine the user acceptance of broadband among the population of households that may be less inclined to adopt newly launched gigabit broadband services. This research paper presents a study about the Urbana-Champaign Big Broadband (UC2B) project; a $29 million BTOP funded middle mile and fiber-to-the-premise project completed in 2013. UC2B is among the first gigabit broadband networks built in the US to provide ultra-high-speed Internet access to the home and community anchor institutions. The network specifically focused on connecting low-income households as well as connecting community anchor institutions that serve households in low-income neighborhoods. The study provides insight into public policy discussions about the public value and return on investment of the construction of gigabit broadband municipal networks. Building the network requires the ability to forecast the demand for the service when potential subscribes face uncertainty about the value of the proposed service. Balking is costly. UC2B, like the other well-known recently built municipal gigabit networks experienced balking. This issue has an economic impact on construction costs and time to delivery, as well as, limits the project’s social impact. This pattern may be particularly prevalent among households located in areas that have been unserved or underserved with broadband. These factors may help to explain why the social and economic impact of broadband investments may be less than expected.
Using data from 2,058 households served by the Urbana-Champaign Big Broadband project (UC2B), the paper examines the adoption of broadband using the Unified Theory for the Acceptance and Use of Technology (UTAUT) (Venkatesh et al., 2003). The model’s determinates of user acceptance and usage behavior includes performance expectancy, effort expectance, social influences and facilitating conditions. The study breaks new ground by considering customer service, consumer response to price and service changes in the local competitive broadband ISP landscape in response to a new ISP, and changing consumer preferences for mobile broadband as additional factors to explain the gap between the intent to adoption broadband and the acceptance and use of broadband.
We collected data for this analysis were collected through mixed method analysis. We are using various approaches to develop a two-stage logistic modeling to estimate the intent of broadband adoption and then to estimate the acceptance and use of broadband adoption. In step 1, we estimate the intent of broadband adoption. In step 2, we estimate the likelihood of broadband adoption as a function of the intent to adopt and other customer service quality factors based on the customer experience between the time the household signed a contract pledging to subscribe to the service and the final successful completion of the service installation at the home.
For the first stage of the model, we estimated the model using data collected through a household survey of computer and Internet use behavior and preferences administered in 2009 and repeated in 2011. The survey asked respondents about the intent to adopt broadband based whether they were likely to subscribe to the UC2B service for $19.99 per month for 20 Mbps service. The responses for the independent variables are measured based on 26 questions using a 5-point Likert scale. The data were collected by a door-to-door canvassing strategy with over 19,000 trips.
For the second stage of the model, we also constructed an adoption indicator using the additional customer service data for households collected through a CRM application capturing each customer interaction indicating, for which households intended to adopt the service by signing a pledge contract, which ones followed through and adopted the service by signing the subscription service contract. Additional independent variables related to the adoption of the broadband service were constructed from an array of data indicating the performance of the construction process of building the middle mile network and the FTTH laterals to connect service to the premise. Factors such as impatience waiting for the construction of the network to be completed, customer service for the installation, intensified competitive responses, or comfort with existing modes of access may turn households away from their decision to adopt broadband. We also are examining data whether the household has existing service from Comcast or AT&T or mobile service and including this in the model. We include changes in competitive pricing and services and preferences for the mode of access including mobile as and alternative that occurred after the new project was announced. These data permit us to extend the UTAUT model to carefully measure customer service of the construction process and competitive factors as an important facilitating condition for the acceptance and use of broadband.
We also collected data through interviews and performed content analysis of the responses to frame the construction of variables for the econometric analysis.
The preliminary econometric modeling to date estimates the intent to adopt the UC2B broadband service. Early results show: Performance expectancy: Measured by the extent to which the household respondent feels that Internet access is important. Performance expectancy is a significant factor in the likelihood that the household intends to adopt broadband. Effort expectancy: Measured by the level of the frequency of use of the Internet. Testing whether being less familiar with using the Internet means that the household is less likely to intend to adopt broadband. The model shows that households that have never used the Internet or are moderate users of the Internet are not likely to intend to adopt. Social Influences: Measured by reporting the extent to which the respondent has friends and family who use the Internet. The model shows that social influences do matter significantly in whether a household intends to adopt broadband. Facilitating Conditions: Measured by has a computer at home is a significant factor. We also found that having children between 5-18 years of age in the household is a significant factor. We included measures to indicate the respondents perception about the affordability of broadband, presence of family members over 60 years of age, perception of Internet security are included in the model but not significant.
Keywords: broadband adoption, user acceptance of broadband, municipal broadband networks, digital inclusion, mixed-methods, broadband network construction
Suggested Citation: Suggested Citation