Method

 < Day Day Up > 



Data

In order to test the model in the context of health information search, an appropriate sample was needed. The Pew Internet and American Life Project commissioned the data collection in the summer of 2000 for a study of online health information. The Pew Project is one of several nonprofit activities that are supported by the Pew Charitable Trusts. The Trusts were created to help organizations and citizens develop practical solutions to difficult problems and have estimated assets in excess of US$3 Billion. Princeton Survey Research Associates (a professional survey company), with support from The Pew Project, collected the data utilized in the present study via a national telephone survey in the U.S. This resulted in a report entitled The Online Health Care Revolution: How the Web helps Americans take better care of themselves (Fox & Rainie, 2000). The data set was made publicly available September 14, 2001. Though the data might be considered somewhat dated if specific websites or strategies were investigated, the constructs under investigation (e.g., search intensity, online behavior) endure as valid research targets, and therefore the dataset fairly captures the research variables of interest. Recall that this study's focus is on the decision process and the role of information acquisition, not purchasing behavior or website atmospherics (Bental & Cawsey, 2002), which may more likely be affected by dated data.

Details of the data collection method, as well as extensive descriptive analysis of the data, can be found in the Pew Organization's report by Fox & Rainie (2000). The initial sample for this study consisted of 521 adult online health information seekers. Subsequently, the data set used here was further refined from the initial sample to include only those that last went online for personal reasons (i.e., looking for information about a health issue concerning themselves), resulting in a final sample dataset of 222 respondents. The final sample allows for the examination of responses that address both the cumulative exposure to online information (long-term effects) as well as the respondent's most recent online health information experience (short-term effects). As such, the data set that was analyzed includes subjects that last went online seeking health information for themselves, their subsequent behaviors during this "snapshot," and how they felt the Internet has affected them overall.

Measurement

To study the hypothesized relationships, the data were examined to determine the items that most represented the constructs of interest. These measures were used to test the hypotheses. The measures utilized to test the model in a health information search context are discussed below, along with descriptive statistics from the 222 respondents.

Independent Variables

Situational Involvement

Situational involvement was measured by a respondent's depiction of their current health status on a 4-point scale. Heath status was ranked excellent (26%), good (57%), fair (14%), or poor (3%). This variable was chosen due to the belief that overall, those that are engaged in an antecedent activity (i.e., poor health) would exhibit a heightened sense of involvement with their health care.

Frequency of Use

Frequency of use was operationalized in a manner similar to Moorthy, Ratchford & Talukdar (1999) and Punj & Staelin (1983) and is expressed as the frequency that an individual uses the Web as a health related information resource. Frequency of use is a cumulative measure in that it assesses the respondent's familiarity with health sites on the web as well as exposure to health information. Frequency of use was obtained in response to the question: How often do you use the Internet to look for advice or information about health or health care: about once a week (30%), once a month (30%), every few months (20%), or less often (20%)?

Mediating Variables

Exploratory Behavior

Hoffman & Novak (1996) depict online exploratory behavior as an individual's engagement in a variety of activities within and across websites. The cumulative behavior of individuals (with respect to health related web-based activities) was evaluated to measure exploratory behavior. The responses to six items were used to form a single measure of exploratory behavior. These six items captured the four factors comprising the quality of the experience: (1) experience gained by purchasing health products and information via the Internet, (2) maintaining an index of frequently visited health information sites, (3) accounting for the source of the health information, and (4) participating in online discussions or chat rooms with individuals concerned with the same health issue. The summed measure was based on learning activities such as: whether the individual has read privacy policies for websites (26% engaged in this activity), kept a site bookmarked (or saved to favorites) (46%), looked to see what organization was providing the information on the website (60%), and online communications that represented responses to whether they had participated in online health support groups (10%), used email to communicate with doctor offices (10%), or described health problems to get advice from an online doctor (8%). Answers to these yes/no questions were summed to form a single measure with a range of zero to six, with six indicating greater levels of exploratory behavior.

Search Intensity

Search intensity assessed the number of health information websites that the subject reported visiting in his or her last online experience. Therefore, as a short-term assessment of search activity, this measure allows us to capture the number of different sources of information the respondent was exposed to during their most recent online session.

Dependent Variables

Life Benefits

The benefit of improved health was assessed through the response to the question: How much, if at all, has getting health and medical information on the Internet improved the way you take care of your health: a lot (15%), some (32%), a little (35%), or not at all (37%).

Communication Effects

The measure of communication effects was assessed by the respondents use of the information obtained the last time they went online for health information—specifically whether or not they spoke to a doctor or nurse about the information they located online (34% did).

Analysis

Multiple regression analysis was conducted to examine the relationships. A number of covariates were also included in order to test for the effects of important demographic characteristics that were influencing Internet adoption at the time of data collection (mainly age, gender, income, and education). Note that the operationalization of communication effects as a binary response variable suggests that logistic analysis is warranted. Logistic analysis and regression analysis resulted in similar findings. The regression results, which are robust to binary dependent variables, are presented here for consistency. The analysis involved three steps, first each of the model variables were regressed onto the covariate set. Next, paths were examined. Finally, mediation models were generated.



 < Day Day Up > 



Advanced Topics in End User Computing (Vol. 3)
Advanced Topics in End User Computing, Vol. 3
ISBN: 1591402573
EAN: 2147483647
Year: 2003
Pages: 191

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net