diffusion in the borderland:hpwren.ucsd.edu/kmb/ica.doc · web viewperhaps the most important...

Diffusion Along the Borderlands: p. 1

Diffusion in the Borderland:

A Study of the Implementation of Broadband

Connectivity in an Ecological Reserve

Tracking Number: ICA-4-10488

Kimberly Mann Bruch (Student)School of Communication–San Diego State University

San Diego Supercomputer Center9500 Gilman Drive, La Jolla, CA 92093-0505

858-822-0977 – [email protected]

Peter A. AndersenSchool of Communication–San Diego State University

5500 Campanile Drive, San Diego, CA [email protected]

Brian SpitzbergSchool of Communication–San Diego State University

5500 Campanile Drive, San Diego, CA [email protected]

Paper submitted for competitive evaluation for presentation at the

International Communication Association Conference,

San Diego, May 2003


Diffusion in the Borderland:

A Study of the Implementation of Broadband

Connectivity in an Ecological Reserve

ABSTRACT

This study investigates the communication and diffusion of a broadband

telecommunications infrastructure (the High Performance Wireless Research and Education

Network - HPWREN). Recently deployed sensors and cameras allow researchers at wildlife

reserves throughout the world to conduct studies and receive sensor and camera data via the

Internet. Based on Rogers’ (1995) diffusion of innovations theoretical framework, field

researchers’ perceptions of network connectivity, communication channels, and use of the

network were assessed. Results supported eight of nine hypotheses related to network

attributes (relative advantage, compatibility, complexity), communication channels, and

network use. Regression analysis of diffusion communication supported the importance of

relative advantage and complexity but not perceived compatibility. Regression analysis of all

variables, with adoption as the dependent variable, showed perceived compatibility,

perceived complexity, and diffusion communication all significantly predict HPWREN

adoption. Results are discussed in terms of the vital role played by communication during the

innovation development, implementation, and use stages.


Diffusion in the Borderland: A Study of the Implementation of BroadbandConnectivity in an Ecological Reserve

Along the borderlands in the extreme southwest of the United States lies the Santa

Margarita Ecological Reserve, part of a pristine network of wildlife reserves located among

the most populous regions of southern California. Located in San Diego and Riverside

counties this 4500-acre ecological reserve is the site for our study of the diffusion of a

National Science Foundation (NSF)-funded high technology broadband information system

called the High Performance Wireless Research and Education Network (HPWREN). This

study examines the diffusion of the HPWREN system among field researchers at the reserve.

THE CONTEXT

The NSF recently proposed that Congress provide federal funding for the

development of a National Ecological Observatories Network (NEON), which first equips

ecological reserves throughout the country with computerized instrumentation (e.g., remote

sensors) and then networks the reserves together (Dalton, 2000). This broadband

telecommunications network would allow for simultaneous transmission of signals (e.g.,

voice, data, video). In turn, both local and national ecologists can easily share field data

collections with one another and also communicate their findings via observatory museums

with targeted publics such as students, the general public, government agencies, and

policymakers. Efforts such as NEON seek to bridge the gap between the scientific

community, the general public, and policymakers.

Scholars such as Valente and Rogers (1995), as well as Dearing, Meyer, and

Kazmierczak (1994), frequently discuss social change and policy research in relation to

technology and scientific innovations. However, a review of more than 100 diffusion studies

reveals dearth of data on diffusion in the scientific community and no quantitative study

regarding the impact of technology upon ecological field research has been conducted. Many

researchers focus their efforts on technology implementation, its use, and adoption among

individuals and organizations (Papa, 1990; Rubinyi, 1989). These studies, however, do not


include specific insights into new communication technologies among scientists or along the

borderlands between ecologists, students in the field of environmental science, applicable

landowners, government agencies, policymakers, and the general public.

Social science research that assesses the current methods of communication among

ecological field researchers, their perceived attitudes toward technology, and the impact of

broadband connectivity upon communication methods and ecological field research will

provide the broader community with a better understanding of how NSF-funded high

technology projects such as the NEON project might positively impact environmental issues

faced by the United States. Creation of high technology systems such as NEON and

HPWREN do not guarantee their success; it is through the process of diffusion of innovations

that humans come to understand, use and benefit from new technologies.

Diffusion of Innovations: Perceived Attributes, Communication, and Adoption

The diffusion of innovations, according to Rogers (1995), is a ubiquitous human

process whereby ideas, processes, and technologies are spread throughout a social system.

He further defines diffusion of innovations as a process involving four elements: (a) one or

more social systems, (b) an innovation, (c) one or more communication channels, and (d)

time. As mentioned, this study examines attributes of a new technological innovation among

field station scientists.

Vital to the adoption of any innovation are five specific attributes: (a) relative

advantage, (b) compatibility, (c) complexity, (d) trialability, and (e) observability (Rogers,

1995). Relative advantage refers to the degree that an innovation appears to be better than its

preceding idea, practice, or object. Relative advantage is typically expressed as the degree to

which an innovation benefits, profits, or generates prestige for the individual or the overall

social system. While the benefits of an innovation may outweigh its costs for one social

system, this may not be the case for other social systems; for example, some individuals

place more value on financial profitability while others may consider other attributes, such as

overall societal benefit, more important.


Similarly, compatibility is the innovation’s level of consistency with previous

experiences, existing ideals, and potential adopters’ needs (Rogers, 1995). Compatibility

encompasses the degree to which an innovation fits with the culture of a social system; for

instance, an innovation developed for one culture may not necessarily be compatible with

another the belief system and needs of another culture. Meanwhile, complexity is how

difficult or easy an innovation appears to an individual (Rogers, 1995). While some

innovations are easy to understand at first glimpse, others are more complicated and require

thorough explanation.

Finally, trialability refers to the degree that an individual can test an innovation

within a limited basis while observability is the visibility level of the innovation–how much

or little others observe the results of an innovation (Rogers, 1995). Related to compatibility,

trialability involves individuals actually trying out an innovation to better understand if and

how it will work for their particular social systems. Likewise, observability allows an

individual to see how an innovation works out for others and hear about the experiences of an

innovation from others. Though trialability and observability are important attributes, the

proposed model specifically focuses on relative advantage, compatibility, and complexity

measures.

Specifically, the diffusion of innovations theory indicates that social systems and

individuals develop innovations when members perceive high relative advantage and

compatibility coupled with low complexity toward discussed innovations (Rogers, 1995). In

other words, members of a social system are less likely to develop innovations if incentives

are not prominent, complexity is high, and there is little relative advantage and compatibility;

whereas members are more likely to adopt an innovation if incentives are prominent,

complexity is low, and perceived relative advantage and compatibility is high. Recently,

Leung and Wei (2000) found that cellular telephones have a high level of compatibility and

relative advantage among individuals, particularly businesspersons, in need of immediate

access to mobile communication. Although this finding is no surprise, their research also


indicated that cell phones provide certain individuals with emotional gratifications, as the

technology is a mechanism for immediate communication with distant family members and

friends; this fits the profile of high relative advantage and compatibility as well. Meanwhile,

cell phones typically have low complexity; therefore, cell phones provide an innovation that

has been cleverly developed for a society that is becoming more mobile and technologically

advanced, yet still needs a method to communicate with family, friends, acquaintances, and

colleagues. Hence, Leung and Wei found high levels of relative advantage and compatibility

among cell phone adopters.

If perceived attributes during the innovation development phase are associated with

adoption individuals should seek information about the innovation through various

communication channels. For instance Rice and Shook (1990) found that positive perceptions

of an innovations attributes was associated with increased communication. Given positive

perceptions about an innovation we suggest that communication will not only increase, it will

be more favorable regarding the innovation. Hence,

H1: As scores within the HPWREN Relative Advantage Index at time one (T1)

increase, favorable multichannelled diffusion communication at time two (T2)

increases.

H2: As scores within the HPWREN Compatibility Index increase (T1), favorable

multichannelled diffusion communication (T2) increases.

H3: As scores within the HPWREN Complexity Index (T1) increase, favorable

multichannelled diffusion communication (T2) decreases.

In turn, favorable communication should be related to positive attitudes toward the

innovation at a later time. Specifically, perceived complexity decreases as social system

members speak more positively and individuals receive favorable mass media exposure

between the innovation development and innovation implementation phases of diffusion.

Likewise, perceived relative advantage and compatibility increases as positive diffusion

communication channels increase. For instance, Valente and Saba (1998) found that mass


media communication channels are positively related to information awareness stages in the

diffusion process while interpersonal communication channels (referred to as “personal

networks” by Valente and Saba) are positively related to behavior change. That is, if a person

has a broad personal network and communicates positively about an innovation, the person

likely perceives the relative advantage and compatibility of an innovation in a more favorable

light; if interpersonal communication channels of an individual are primarily negative, the

person likely perceives the innovation’s relative advantage and compatibility in a more

negative light. On the other hand, “individuals who lack personal contact with many users of

an innovation may turn to the mass media for information about new ideas and products” (p.

116). Hence it is likely that positive communication is associated with more positive

attributes toward key elements of an innovation.

H4: As positive diffusion communication channels at time 2 (T2) increase, the

degree of perceived complexity (T2) decreases.

H5: As positive diffusion communication channels (T2) increase, the degree of

perceived relative advantage (T2) increases.

H6: As positive diffusion communication channels (T2) increase, the degree of

perceived compatibility (T2) increases.

The proposed model suggests that diffusion communication coupled with continued

perceptions eventually lead to the adoption decision: an individual either adopts or rejects the

innovation (see Figure 1).

As previously discussed, Rogers (1995) found that positive relationships exist

between relative advantage and adoption as well as compatibility and adoption. Meanwhile,

Rogers illustrated a negative relationship between complexity and adoption rates. Additional

diffusion studies confirm the relationships posited by Rogers. For instance, Rubinyi (1989)

examined an office setting originally unfamiliar with computerized systems and their efforts

to computerize all internal office functions. After one year, more than 90% of the groups did

not have the majority of their network functions up and running. However, by the end of the


two-year study, most of the groups were networking data between their own group and other

groups. On the other hand, computer use in general did not have the same slow adoption rate.

Rubinyi attributed the slow networking adoption rate to its complexity–compared with

simply using computers, networking takes more coordination with others and more technical

savvy than simply operating a single computer that is not networked to others.

Likewise, Manross and Rice (1986) found that technical savvy, or lack thereof, also

played a large role in the adoption rate of an intelligent telephone in an organization.

Although their study did not find a correlation between overall attitudes and usage of the

phone’s specific functions, Manross and Rice discovered a difference between the

relationship among those users at differing organizational levels (e.g., managers versus

administrative assistants), their attitudes, and adoption decisions. Further, Kleinman (2000)

found that the flexibility of CMC provided high relative advantage among women scientists

and engineers who joined a 24-hour on-line forum, which allowed them to share their

research with peers and mentors. Consistent with this research we posit that:


H7: Subjects who score high within the HPWREN Relative Advantage Index at time

2 (T2) are more likely to use the network connectivity than those who score

low.

H8: Subjects who score high within the HPWREN Compatibility Index (T2) are

more likely to use the network connectivity than those who score low.

H9: Subjects who score high within the HPWREN Complexity Index (T2) are less

likely to use the network connectivity than those who score low.

METHODS

Participants

The participants were field scientists, ranging from evolutionary biologists to wildfire

management researchers, affiliated with Field Stations Program of a major Southwestern

University that oversees the Santa Margarita Ecological Reserve. Thirty-seven of the 40

eligible field scientists participated in the study (a 92.5 % participation rate). The respondents

were 81% male and 19% female, 54% are faculty members from southern California

universities, 41% are staff at academic, military, and research institutions and agencies and

5% are graduate students at the aforementioned University.

Setting

Spanning approximately 4500 acres with elevation ranges from 150 to 2300 feet, the

Santa Margarita Ecological Reserve is nestled between the Santa Ana Mountains in the

northeast portion of San Diego County. Home to one of the last free-flowing streams in

southern California, topography is rugged and complex with low hills and intervening

drainages. In Fall 2001, HPWREN researchers began development and implementation of a

broadband telecommunications infrastructure and reserve field scientists now have the

capability of sending and receiving data at a speed of 45 megabits per second (Figure 2).


More than 50 research projects are being conducted at reserve - including a wide

array of studies, such as (a) threatened and endangered species; (b) water quality and public

health; (c) agriculture; (d) global change; (e) fundamental ecology, geology and geography;

and (f) environmental engineering. The newly established high-speed data connection,

however, prompts additional researchers to utilize technology as a means of data collection

and transmission. For instance, a weather station (Figure 3) represents only one example of

the many sensors being implemented throughout the reserve.


The reserve is being equipped with additional sensors that provide researchers with

high-resolution time series measurements of physical, chemical, and biological variables

found not only in the air, but also in the watershed. Systems that capture both audio and

video are also being deployed in the gorge of the reserve so that scientists can remotely

monitor habitats (Figure 4). Consequently, ecoinformatics development such as this falls

right in line with the NSF’s recently proposed National Ecological Observatories Network

(NEON). With the goal of equipping ecological reserves around the country with

computerized instrumentation infrastructures, NEON would also network all of the reserves

together via broadband telecommunications.

As mentioned in the introduction of this paper, the proposed NEON project would

allow U.S. ecologists to easily share field data collections with one another and communicate

their findings with their students, government agencies, private landowners, and the general

public.


Procedures

Permission to collect participant data regarding broadband connectivity use, diffusion

communication, and relative advantage, compatibility, and complexity was granted to

researchers, including the author, affiliated with the NSF-funded High Performance Wireless

Research and Education Network (HPWREN) by Human Subjects Committees at two major

Southern California Universities. Pre-connectivity data were collected from HPWREN

researchers in December 2001 while broadband connectivity via HPWREN was being

established at the reserve; post-connectivity data were collected in March 2002. Telephone

interviews were used to collect diffusion communication, relative advantage, compatibility,

and complexity data and email surveys were used to collect HPWREN usage data.

Specifically, the variables were measured using two sets of questions: the pre-test

(see Appendix A) was conducted prior to HPWREN connectivity while the post-test (see

Appendix B) was administered after connectivity was established. Pre-test survey questions

concerned perceived attributes and were administered at time one. The pre-test survey


consisted of 24 items related to perceived relative advantage, compatibility, and complexity;

this telephone survey was administered during the innovation development stage.

Subsequently, post-test surveys consisted of 27 items and were conducted in two phases. The

first 21 questions (time two) examined continued perceptions (relative advantage,

compatibility and complexity) while the final six questions (time three) measured adoption

decision. The survey questions were developed based upon prior research conducted by

Rogers (1995), Valente (1996), Rice (1994), Rice and Gattiker (2001), Spitzberg (2002), and

Reagan (1989).

In addition to quantitative data collection specific to the hypotheses posited in this

thesis, qualitative data were also recorded. This information provided insight regarding the

specific real-time data that adopters plan to measure, their use of specialized sensors, needed

technical support, and HPWREN training that users consider beneficial. Additionally,

qualitative data allowed for a better understanding of the non-adopters’ decisions to reject the

innovation.

Variables

Using the aforementioned diffusion of innovations theoretical framework (Rogers,

1995), the dependent variable of this research was the adoption of the innovation: broadband

connectivity. The independent variables consisted of perceived attributes (relative advantage,

compatibility, and complexity) as well as diffusion communication–though both

interpersonal and mass media channels.

Perceived Attributes and Continued Perceptions

Perceived attributes, which were measured during pre-test surveys, refer to the

attitudes during the innovation development. Continued perceptions, which were measured

during the post-test surveys, refer to the attitudes during the innovation implementation and

use. Because the field scientists were not heavily involved with the development process,

however, a significant difference between perceptions at time one and time two surveys was

not anticipated. On the other hand, comparison of pre-connectivity perceptions with post-


connectivity perceptions proves interesting for HPWREN researchers as they examine the

evolution of understanding among broadband telecommunications users and provide the most

efficient infrastructure and training materials. Specifically, several indexes were used to

measure relative advantage, compatibility, and complexity.

Relative Advantage. To measure relative advantage, five items were employed: the

extent to which HPWREN (a) improves the field scientist’s research endeavors; (b) improves

teaching efforts; (c) allows for more efficient information dissemination; (d) benefits

outweigh costs; and (e) has more advantages than disadvantages. Cronbach reliability

estimates indicated that the relative advantage measure was reliable (T1 = .81; T2 = .84).

Compatibility. Five items were employed that ascertained HPWREN’s compatibility

with the participant’s (a) current technology; (b) level of network training; (c) research

needs; (d) methods of data collection; and (e) teaching curricula. Cronbach reliability

estimates indicated the five-item version of the compatibility index was moderately reliable

(T1= .73; T2 = .71).

Complexity. Similarly, five measured complexity by determining whether or not the

participant (a) has questions about HPWREN; (b) perceives the network as difficult to use;

(c) must acquire technical assistance to use HPWREN; (d) prefers person-to-person technical

assistance; and (e) is not very knowledgeable about networking technology. Cronbach

reliability estimates indicated that substantial increase in reliability for the complexity

measure was achieved when one item was removed from the index, so the item measuring

preference for person-to-person technical assistance was suggested for deletion from the

complexity index resulting in a four-item version of the complexity index which was

moderately reliable (T1 = .70; T2 = .74).

Diffusion Communication

Diffusion communication encompasses both interpersonal and mass media

communication channels. Scholars measure interpersonal communication channels using

Likert-scale questions similar to those in this study (e.g., Rice, 1991; Valente & Saba, 1998).


To measure interpersonal communication channels specific to innovation perceptions, the

participants were asked about the extent to which information they received about HPWREN

was favorable: (a) within their workplace and (b) outside of their workplace. To measure

mass media communication channels specific to innovation perceptions, the participants

were asked about the extent to which they received favorable information about HPWREN:

(a) in print media; (b) broadcast media; and (c) electronic media (see Appendix B). The

highest level of reliability was attained by combining these measures to form a single

diffusion communication measure ( = .68).

Adoption Decision

In order to better understand the adoption decision of participants, they were asked

about their (a) current use of the network; (b) perceived use in their future work; (c)

perceived future use for research endeavors; (d) perceived future use for teaching efforts; and

(e) perceived future use for overall purposes. These variables were measured using standard

attitude scales on part and future frequency of use that were administered using electronic

mail surveys (see Appendix B, questions 22-26). Cronbach reliability analysis indicated that

the five measures for adoption decision were all reliable ( = .88).

RESULTS

A path model (Figure 1) was developed to provide an omnibus test of the hypotheses.

Given the size of the sample, true multivariate path analysis was considered inappropriate.

Instead, bivariate correlations and multiple regression were used to test the model.

Hypothesis Tests

Eight of the nine posited hypotheses were supported using bivariate correlations

(Figure 5). Hypothesis one, which predicted that as scores within the HPWREN Relative

Advantage Index (T1) increase, positive diffusion communication channels (T2) increase,

was supported (r = .56, p < .001, r2 = .31). Likewise, hypothesis two, which stated as scores

within the HPWREN Compatibility Index (T1) increase, positive diffusion communication

channels (T2) increase, was also supported (r = .52, p < .001, r2 = .27). Similarly, as the initial


perceived complexity of HPWREN increased among study participants, positive diffusion

communication channels decreased and hypothesis three, which stated as scores within the

HPWREN Complexity Index (T1) increase, positive diffusion communication channels (T2)

decrease, was also supported (r = -.44, p = .007, r2 = .19).

Hypothesis four, which stated as positive diffusion communication channels (T2)

decreases, the degree of perceived complexity (T2) decreases, was supported (r = -.52, p

= .001, r2 = .27). Likewise, hypothesis five, predicting that as positive diffusion

communication channels (T2) increase, perceived relative advantage (T2) increases, was

supported (r = .57, p < .001, r2 = .33). Additionally, hypothesis six, which anticipated that as

positive diffusion communication channels (T2) increase the degree of perceived

compatibility (T2) increases, was also supported (r = .48, p = .003, r2 = .23).

Finally, as the continued perceptions regarding relative advantage and compatibility

of HPWREN increased among study participants, innovation acceptance increased. That is,

as stated in hypotheses seven and eight, subjects who score high within the HPWREN

Relative Advantage Index (T2) are more likely to use the network connectivity (T3) than

those who score low (r = .68, p < .001, r2 = .46) and subjects who score high within the

HPWREN Compatibility Index (T2) are more likely to use the network connectivity (T3)

than those who score low (r = .55, p < .001, r2 = .30). However, as the continued perceptions

regarding complexity of HPWREN increased, innovation acceptance did not decrease.

Therefore, hypotheses one through eight were supported, while hypothesis nine,

which predicted that subjects who score high within the HPWREN Complexity Index (T2)

are less likely to use the network connectivity (T3) than those who score low was not

supported (r = -.31, p = .066).


As an exploratory analysis, paired samples t-tests were performed to detect

differences in the means for relative advantage, compatibility, and complexity. The t-tests

showed no overall valence shift between perceived attributes during the innovation

development stage and continued perceptions during the innovation implementation and use

stage.

Multiple Regression

Again, given the small sample, it was considered inappropriate to conduct a complete

multivariate path analysis or structural equation modeling. Therefore, to further test the

analytic model for direct effects and mediator variables, hierarchical multiple regression was

performed. The procedure was to enter groups of variables in the model first, and then the

next, and so forth. The sequence was then altered systematically. This procedure permitted an

interpretation of moderation or mediation among the variables (Baron & Kenny, 1986).

Then, the best fitting model given these results was assembled. Recognizing that there is

substantial room for interpretation in such a procedure, the resulting model (Figure 6)

indicated that initial perceptions of attributes (relative advantage and complexity at time one)

impact diffusion communication related to the innovation (HPWREN). Further, initial

perceptions of attributes (compatibility and complexity at time one) and diffusion

communication predict innovation adoption. However, continued perceptions of attributes

(relative advantage, compatibility, and complexity at time two) were not shown to be

significant predictors of innovation adoption. These quantitative results, as well as qualitative

results, are explicated in the discussion section of this paper.

DISCUSSION

Results using bivariate correlations provided support for eight of the nine posited

hypotheses. In the discussion we present some qualitative data that provides additional

insight regarding why these hypotheses were supported. Finally additional analysis using

multiple regression provided a more complete test of the entire diffusion model.


Hypotheses One and Two

Strong positive relationships between both relative advantage and compatability with

positive diffusion communication channels suggest that these initial positive predispositions

lead to the seeking of positive information about the innovation. These two variables, which

each explain 31% and 27% of the variance respectively in positive diffusion communication,

are exemplified by the following qualitative data collected during T1 interviews:

“Basically, we have a couple of people with dip nets that examine fish spawn. What we do is hike up the stream, take multiple samples, and count the number of fish - determining species and other things. We then type our field notes directly into a PDA and we actually hope to implement field records into an Internet database soon. Yes, real-time digital field notes would be terrific. I admire SDSU for pursuing this, as I find it entirely appropriate.” -Fish ecologist

“I’ve been talking with faculty around the country about using real-time data in their classrooms and feedback has been very positive.” -Biology professor

Hypothesis Three

Results showed that complexity is strongly and negatively related to positive

communication about the innovation with complexity explaining about 27% of the variance

in positive diffusion communication. These results support the notion that complexity make

potential adopters fail to seek information or find negative messages about an innovation.

These relationships can best be better understood by the following qualitative data collected

during T1 interviews.

“Monitoring via sensors and equipment takes maintenance, which can be difficult to implement without causing disturbance...I did hear that this network was in the plan for the reserve, but don’t know anything about it.” -Golden eagle researcher

“The automated collection of biological data is very hard...Broadband is wonderful to have there, but the biggest question is how do we automate biological sensing? Field biological collection tools are not yet to the point to fully utilize this type of technology. I just haven’t really heard much about the technical aspects of this network, so I can’t really say that much right now.” -Mammologist

“The USGS has streamflow data available online and it is organized really well with a good interface. It is simple, easy to use...Organization of data can impact


the usefulness of the data...Ease of access will control the usage of this data, too.” -Watershed researcher

Hypothesis Four

Results of the present study showed that as positive diffusion communication

channels increase, the degree of perceived complexity substantially decreases. Over one

quarter of the variance in reduced complexity is a function of positive diffusion

communication. Alternatively, this hypothesis also suggests that negative communication

increases the perceived complexity. Qualitative data typify this relationship.

“It would be nice to keep everyone up to date about the network - to keep all in touch with available data...Keep in touch with the Supercomputer Center and see how we can best incorporate multiple databases, because that is what they do there and would help us know more about the network.” -Biology professor

Hypotheses Five and Six

Results showed that as positive diffusion communication channels increase, the

degree of perceived relative advantage and compatibility increases. Positive communication

explains about 1/3 of the variance in relative advantage and nearly a quarter of the variance

in compatibility. The following qualitative data collected during T2 interviews with

respondents provides additional data about this process:

“I just received your packet of information about the network and am now taking a class out to SMER...I hope to show them some of this technology out in the field.” -Biology professor

“The network has just been explained to me by SMER staff and I would love to use this sometime in the future. Before when you talked to me, I just didn’t know enough about it, but now I may try to work out some research project around the technology in the future.” -Mammologist

These results also reveal that respondents who have negative, or simply a lack of,

diffusion communication channels will not have a positive perception regarding relative

advantage and compatibility. Instead, these respondents are likely to have a negative or no

opinion about the network’s relative advantage and compatibility with their work:


“I am still not sure about HPWREN, but that is probably because I have not heard about it as I am somewhat disjunct from the school as an adjunct faculty member.” -Adjunct biology professor

Hypotheses Seven and Eight

Perhaps the most important findings for diffusion theory were that subjects who

perceive HPWREN had high relative advantage and compatibility were more likely to use it

than those who score low. Relative advantage explained 46% of the variance and

compatibility explained 30% of the variance in adoption.

The qualitative data reveal some of the nuances of this process:

“I will soon be recording sounds from owls, as there may be a correlation between moon phase and weather. Real-time weather data streamed via this network and correlated with owl data and moon phase data will allow me to measure the spacing between calls and see if there are relationships between the sounds, weather, and moon phases...Yes, I will use this network.” -Biologist

“We now have our moss research at the James Reserve online and it’s great...We want to set up more of these real-time stations...We need more experiments to gauge results...Yes, we will use HPWREN at SMER.” -Fungus and moss researcher

“This is a fantastic program and I’m anxious to see how this changes both research and teaching in profound ways. I’d also be interested in seeing the use of SGI Visual Area Networks with HPWREN, as I am already using both networks.” -Geology professor

Likewise, respondents who score lower within the relative advantage and

compatibility indices are less likely to adopt the technological innovation. Further qualitative

results, such as the following comment from T2 interviews, support this hypothesis.

“My biggest concern is not the technology, but HPWREN is just not going to really aid me with my data collection. This would require radio collars on the animals, which to my knowledge is not easily available at this time. Although there are passive integrative transponders that allow for the animal to be injected with a tag under the skin and then records their actions as they move through underground hoop antennas, this technique requires hoop antennas everywhere - which causes a trampling issue. There are also radio transmitters that you can put in an animal...this is used largely in quail research...the animal transmits a signal


to a nearby radio tower. However, this technique is rarely used due to challenges like automatic triangulation. If auto-triangulation was in place at SMER, we could use this, but until then, it is just not feasible for my particular research agenda.” -Mammologist

Hypothesis Nine

The last hypothesis predicted that subjects scoring higher on the complexity index at

time 2 would be less likely to use the network connectivity than those who score low on the

complexity index. Hypothesis nine was not supported. Though results were in the

hypothesized direction in that subjects who thought HPWREN was highly complex, were

less likely to use the system than subjects who believed it was fairly simple the relationship.

However, the relationship was not statistically significant.

Because simplicity typically attracts more users to a technological innovation than

complexity (Rogers, 1995), this finding was somewhat surprising. However, it should be

noted that the current sample consists of highly educated field scientists who are possibly less

intimidated by “high-tech” innovations than less educated adopters. Further, multiple

regression results contradicted the bivariate correlations and showed that perceived

complexity (T1) has an impact upon both diffusion communication (T2) and adoption (T3);

see Figure 6 for details.


Evaluating the Overall Model

Simultaneous multiple regression analysis of diffusion communication with relative

advantage, compatibility, and complexity as predictors resulted in significance for both

relative advantage (r = .55, p = .02, r2 = .30) and complexity (r = .36, p = .029, r2

= .13). That

is, diffusion communication (T2) is significantly affected by perceived relative advantage

(T1) and perceived complexity (T1); however, multiple regression results indicated that

perceived compatibility is not a significant predictor of diffusion communication (see Figure

6). However, when adoption is viewed as a dependent variable for the time 1 variables of

relative advantage, compatibility, complexity, and the time 2 variable of diffusion

communication, compatibility reappears to play a role in predicting adoption (r = .45, p

= .04, r2 = 20) along with perceived complexity (r = .37, p = .04, r2 = .14) and diffusion

communication (r = .33, p = .04, r2 = .11). However, relative advantage at time 1 falls out of

the picture. Whereas relative advantage at time 2 is strongly related to adoption at time 2 (r

= .68, p < .001, r2 = .46), at time 1 its influence appears to be mediated by other factors.

Perhaps the most important lesson learned during this study was the vital role played

by communication during the innovation development, implementation, and use stages. The

interview process alone allowed many field scientists at the reserve to consider an innovation

(HPWREN) that they might have otherwise never heard about. Further, the study showed

that even though some respondents are not technologically savvy, they are willing to adopt

technological innovations - provided perceived attributes are favorable.

Although a few respondents already use PDAs, the majority of respondents use a

manual field notebook for recording data collection findings. They then transfer the

information to their office or laboratory computer. However, qualitative data, such as the

following, indicates consideration of new electronic field devices:

“It depends on where I am going, as most of the time it is easier to carry a notebook then a computer in remote areas. I am thinking about getting a PDA though.” -Hydrologist


Again, this type of insight shows that if the innovation proves to be compatible with

their current data collection procedures, the field researcher is likely to utilize it - as long as

they receive information about it via their communication channels.

Limitations of the Study

Even though this study proved to be quite successful, at least three limitations are

apparent: (a) only one reserve was examined, (b) only one innovation was studied, and (c)

the sample was relatively small. In order to compare the results of the Santa Margarita

Ecological Reserve study participants with another sample, broadband connectivity must first

be established at an additional reserve. This would allow for further analysis of the

hypotheses. Consequently, HPWREN researchers are currently planning to connect the

nearby Sky Oaks Field Station to the network; this will provide an optimal comparative

sample set.

The current study’s limitation of only one innovation was also somewhat problematic.

When only generalizing to the adoption of HPWREN, the diffusion of innovations theoretical

framework works perfectly and results are impressive; however, if generalizing to overall

innovation diffusion, this study is somewhat problematic as only one innovation was

examined - rather than multiple innovations. Therefore, because the research examines only

one innovation, HPWREN, the results cannot be generalized to all technological innovations

(e.g., data-loggers and PDAs) related to field science. Perhaps another study that examines

the attitudes, communication, and adoptions of data-loggers and PDAs among field scientists

would also prove useful for the NSF and additional entities.

Finally, the sample size was also somewhat limiting for the study. Although the

sample consisted of nearly all (37 of 40) field scientists affiliated with the Santa Margarita

Ecological Reserve, this small number of participants is somewhat problematic. Again, an

additional study that examines Sky Oaks scientists would help address the issue.


Future Implications

This research not only quantitatively describes the attitudes of the field researchers

toward broadband connectivity, but also outlines a model for a longitudinal future study.

From the analytical research standpoint, the described objectives are important, as additional

ecological reserves may be added to the NSF-funded HPWREN and diffusion of this

technology is better understood by determining the perceived attributes and utility of

broadband connectivity among ecological field researchers–as oftentimes these perceptions

are discussed in qualitative manners, but never truly quantified. Additionally, this study

provides a propositional model that may be used for future longitudinal studies involving the

diffusion of technological innovations among particular social systems.

Perhaps even more importantly, the study has the potential for providing quantitative

information to an array of publics regarding the impact of broadband connectivity on

ecological research. These findings further allow policymakers, such as those charged with

NSF proposals like NEON, to better understand how such environmental monitoring projects

might positively impact the future of ecological monitoring and the overall environmental

research agenda in the United States.

Because our society often relates environmental protection with technological

innovations (Miller & Garnsey, 2000), it is important that scholars continue to research the

diffusion process and publish their findings in applicable journals so that the information is

widely disseminated. Further, diffusion studies that better explain the benefits to the

environment via particular technological innovations should be a focus of mass media

channels so that the general public is also aware of important environmental issues and how

their attitudes and behaviors impact the environment.

Additionally, NEON researchers may be able to learn from the Dutch science shops,

which have existed for more than 30 years (Farkas, 1999), and better match the agenda of

environmental research with that of the greater community by providing not only information


about the issues and concerns at hand, but immediate ways that individuals can participate in

science and environmental protection initiatives.

Further, initiatives like NEON also allow societal members other than those with the

most education and financial resources to access information that is typically reserved for

citizens that are the best-educated and most wealthy (Miller & Garnsey, 2000). That is, once

NEON sites are in place, common citizens can virtually tour ecological observatories around

the country, ranging from the mountaintops of Colorado’s front range to the swamplands of

the Mississippi delta. NEON will not only be a resource that allows scientists to better

communicate with one another and thereby promote interdisciplinary collaborations, but the

network has the potential of serving as a national treasure to be shared by every citizen.


REFERENCES

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Dalton, R. (2000). NEON to shed light on environment research. Nature, 404, 216.

Dearing, J. W., Meyer, G., & Kazmierczak, J. (1994). Portraying the new: Communication between university innovators and potential users. Science Communication, 16, 11-42.

Farkas, N. (1999). Dutch science shops: Matching community needs with university R&D. Science Studies, 2, 256-273.

Kleinman, S. S. (2000). Social identification in a computer-mediated group for women in science and engineering. Science Communication, 21, 344-366.

Leung, L., & Wei, R. (2000). More than just talk on the move: Uses and gratifications of the cellular phone. Journalism and Mass Communication Quarterly, 77, 308-320.

Manross, G. G., & Rice, R. E. (1986). Don’t hang up: Organizational diffusion of the intelligent telephone. North Holland Information & Management, 10, 161-175.

Miller, D., & Garnsey, E. (2000). Entrepreneurs and technology diffusion: How diffusion research can benefit from a greater understanding of entrepreneurship. Technology in Society, 22, 445-465.

Papa, M. J. (1990). Communication network patterns and employee performance with new technology. Communication Research, 17, 344-368.

Reagan, J. (1989). New technologies and news use: Adopters vs. nonadopters. Journalism Quarterly, 66, 871-877.

Rice, R. E. (1991). Organic organizations and centralized units: Use, contexts, and outcomes of word processing. In J. Morell & M. Fleischer (Eds.), Advances in implementation and impact of computer systems (pp. 111-141.). New York: JAI.

Rice, R. E. (1994). Network analysis and computer-mediated communication systems. In S. Wasserman & J. Galaskiewicz (Eds.) Advances in social network analysis (pp. 167-203). Newbury Park, CA: Sage.

Rice, R. E., & Gattiker, V. E. (2001). New media and organizational structuring. In F. Jablin & L. Putman (Eds.), The new handbook of organizational communication: Advances in theory, research, and methods (pp. 544-581). Thousand Oaks, CA: Sage.


Rice, R. E., & Shook, D. E. (1990). Voice messaging, coordination, and communication. In J. Galegher, R. Kraut, & C. Egido (Eds.), Intellectual teamwork: Social and technology foundation of cooperative work (pp. 327-350). Hillsdale, NJ: Lawrence Erlbaum.

Rogers, E. M. (1995). Diffusion of innovations (4th ed.). New York: Free Press.

Rubinyi, R. M. (1989). Computers and community: The organizational impact. Journal of Communication, 39, 110-123.

Spitzberg, B. (2002). Preliminary Development of a Model and Measure of Computer Mediated Communication (CMC) Competence. Unpublished manuscript, San Diego State University.

Valente, T. W. (1996). Social network thresholds in the diffusion of innovations. Social Networks, 18, 69-89.

Valente, T. W., & Rogers, E. M. (1995). The origins and development of the diffusion of innovations paradigm as an example of scientific growth. Science Communication, 16, 242-273.

Valente, T. W., & Saba, W. P. (1998). Mass media and interpersonal influence in a reproductive health communication campaign in Bolivia. Communication Research, 25, 96-124.


APPENDIX A

PRE-TEST SURVEY QUESTIONS


TELEPHONE QUESTIONNAIRE SCRIPT< > indicates note to interviewer only - not to be read aloud to subject.

<1. Enter phone number as ID variable.><In the case of missing data, indicate 9 in the applicable edge coding section.>

Hi. I’m a graduate student and am following up on an email that I sent you a few days ago regarding an NSF-funded research project that is providing Santa Margarita Ecological Reserve and the Sky Oaks Field Station with high-speed network connectivity.

I’m calling to ask you a few questions about the impact of the network upon your research and your teaching, and which future networking schemes could best benefit your work.

Basically, the two Field Stations are currently being connected to a high-speed network called HPWREN, which is funded by the National Science Foundation. The speed of the network is around 45 Megabits per second and will allow researchers like you to transmit large amounts of real-time information from the field station to the campus database where it can be readily accessed by researchers and students through a website. An in-field station network of automated data acquisition towers are also being developed; these towers are intended to assist researchers like you with data collection methods.

As mentioned in my email, I’d like to ask you a few questions about the impact of this network connectivity upon your research.

2. What will you and your team need to utilize the network? (Answer yes to as many of these as applicable). Do you need...

a. computer equipment b. specialized sensorsc. technical assistance/support (e.g., technician)d. traininge. other: ______________________________

3. What are your sources of data for your field experiments? (Again, answer yes to as many of these as applicable). Do you use...

a. soil samples from plots b.climate data from towerc. other: ________________________________

4. What are your current datalogging methods? Do you...a. write in field book, data forms, or maps, and later type into computer (computer brand/model: _______________)b. type directly into PDA from field (PDA brand/model:___________________)c. laptop computer (laptop brand/model:___________________) d. other: ______________________________

5. Does your research...a. Alwaysb. Sometimesc. Seldom or d. Never

...require continuous measurements (continuous = measurements that are taken more than once per day)?


6. What is your sample interval (how often)?

7. Do you...a. Alwaysb. Sometimesc. Seldom or d. Never

...use specific sensors (e.g., cameras, meteorological data, soil probes, etc)?

8. Describe the sensors that you use: past, present, and hopefully future (i.e., what types of data could be useful for you in the future)?

Okay, that sounds really interesting. I’ll make sure and relay that information to Sedra and Hans-Werner. Now, I have a few more questions for you. If I were putting words into your mouth about the following statements, would you strongly agree, agree, have no opinion, disagree, or strongly disagree....

9. HPWREN will help me improve my research endeavors by providing me with better access to real-time data than my current system. Do you...

a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)

...with that statement?

10. HPWREN will help me improve my teaching efforts by providing me with improved access to real-time data in the classroom setting.


11. HPWREN will allow me to disseminate research information to colleagues and applicable publics in a more efficient manner.


12. All things considered, HPWREN benefits outweigh costs.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)


13. Overall, HPWREN has more advantages than disadvantages.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)

14. HPWREN would be very compatible with my current technology.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)

15. I will need little specialized training to use HPWREN technology.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)

16. HPWREN meets several technological needs of my research project.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)

17. I will be able to use HPWREN for research purposes, without changing the structure of my field data collection procedures.


18. I will be able to use HPWREN for teaching purposes, without significantly changing the structure of my course curricula.


19. I have questions about the basic concepts of HPWREN and how real-time field data collection and transmission works.



20. HPWREN will be somewhat difficult for my staff and me to use for data collection and transmission.


21. I will have to acquire the help of a technical person to help me with the use of HPWREN.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)e. Strongly disagree (1)

22. When I have a problem with a technological device, I prefer person-to-person technical assistance.


23. I am not very knowledgeable about technology and do not currently feel comfortable using HPWREN.


Okay, I just have one final question for you...

24. In the context of measuring the impact of broadband connectivity upon your field research, can you think of other things you would like to communicate that I should have asked you about? Or, do you have any comments that you would like to add for my study?

Once the network connectivity is in place, I will be contacting you for a post-test evaluation. I really appreciate your time. Thank you.


APPENDIX B

POST-TEST SURVEY QUESTIONS


TELEPHONE QUESTIONNAIRE SCRIPT< > indicates note to interviewer only - not to be read aloud to subject.<Enter phone number as ID variable.><In the case of missing data, indicate 9 in the applicable edge coding section.>

Hi. I’m a graduate student and am following up with the additional survey that I mentioned to you regarding an NSF-funded research project that is providing Santa Margarita Ecological Reserve and the Sky Oaks Field Station with high-speed network connectivity. I’m calling to ask you a few more questions about the impact of the network upon your research and your teaching, and which future networking schemes could best benefit your work. As you know, the Santa Margarita Ecological Reserve has been connected to a high-speed network called HPWREN, which is funded by the National Science Foundation. The speed of the network is around 45 Megabits per second and allows researchers like you to transmit large amounts of real-time information from the field station to the campus database where it can be readily accessed by researchers and students through a website. An in-field station network of automated data acquisition towers continue to be developed; these towers are intended to assist researchers like you with data collection methods. Just like the last time we talked, I have a few questions about the impact of this network connectivity upon your research.

1. First of all, did you receive the packet of information that I sent you via the mail last week?a. Yes (2)b. No (1)

Okay, now, here are a few more questions that I have for you. They will seem familiar, because I asked you most of these questions just a couple of months ago. If I were putting words into your mouth about the following statements, would you strongly agree, agree, have no opinion, disagree, or strongly disagree....

2. HPWREN will help me improve my research endeavors by providing me with better access to real-time data than my current system. Do you...

a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1) ...with that statement?

3. HPWREN will help me improve my teaching efforts by providing me with improved access to real-time data in the classroom setting.

a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

4. HPWREN will allow me to disseminate research information to colleagues and applicable publics in a more efficient manner.

a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)


e. Strongly disagree (1)

5. All things considered, HPWREN benefits outweigh the costs.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

6. Overall, HPWREN has more advantages than disadvantages.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

7. HPWREN would be very compatible with my current technology.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

8. I will need little specialized training to use HPWREN technology.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

9. HPWREN meets several technological needs of my research project.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

10. I will be able to use HPWREN for research purposes, without changing the structure of my field data collection procedures.


11. I will be able to use HPWREN for teaching purposes, without significantly changing the structure of my course curricula.

a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2)


e. Strongly disagree (1)12. I have questions about the basic concepts of HPWREN and how real-time field data collection and transmission works.


13. HPWREN will be somewhat difficult for my staff and me to use for data collection and transmission.


14. I will have to acquire the help of a technical person to help me with the use of HPWREN.a. Strongly agree (5)b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

15. When I have a problem with a technological device, I prefer person-to-person technical assistance.


16. I am not very knowledgeable about technology and do not currently feel comfortable using HPWREN.


17. The information that I have received and heard about HPWREN from colleagues within my workplace has been favorable.


18. The information that I have received about HPWREN from acquaintances outside of my workplace has been favorable.

a. Strongly agree (5)


b. Agree (4)c. Have no opinion (3)d. Disagree, or (2) e. Strongly disagree (1)

19. The information that I have received and heard about HPWREN from print media (e.g., newspapers, magazines) has been favorable.


20. The information that I have received and heard about HPWREN from broadcast media (e.g., television, radio) has been favorable.


21. The information that I have received and heard about HPWREN from electronic media (e.g., websites, electronic newsletters) has been favorable.


22. Are you currently using HPWREN?a. Yes (2)b. No (1)

23. Will you use HPWREN in your future work?a. Yes - Definitely (5)b. Yes - Probably - (4)c. Unsure (3)d. No - Probably (2)e. No - Definitely (1)

24. I am likely to use HPWREN in my research in the coming year.a. Strongly agree (5)b. Agree (4) c. Have no opinion (3) d. Disagree (2)e. Strongly disagree (1)

25. I am likely to use HPWREN in my teaching in the coming year.a. Strongly agree (5)b. Agree (4)


c. Have no opinion (3) d. Disagree (2)e. Strongly disagree (1)

26. I am likely to use HPWREN in some way in my work in the coming year.a. Strongly agree (5)b. Agree (4) c. Have no opinion (3) d. Disagree (2)e. Strongly disagree (1)

Okay, I just have one final question for you... 27. In the context of measuring the impact of broadband connectivity upon your field research, can you think of other things you would like to communicate that I should have asked you about? Or, do you have any comments that you would like to add for my study?

diffusion in the borderland:hpwren.ucsd.edu/kmb/ica.doc · web viewperhaps the most important...

Documents