The questionnaire was distributed using mailed surveys, the world wide web (WWW), e-text (e-mail) and telephone interviews. Of the respondents, 89% used the WWW (n = 141), 6% used the printed questionnaire (n = 10), 3% requested a telephone interview (n = 4), and 2% used e-text (n = 3). Although access to the Web is not universal, it is a very appropriate technology for gathering information from persons with spinal cord injuries. Currently, the frequent practice is to train persons with a spinal cord injury to be computer literate early in their Rehabilitation.
Persons using Internet access responded via the WWW page from the continents of Asia, Africa, Europe and North America. One hundred and forty-three (90.5%) of the participants identified the United States as their country of residence. Persons from 36 of the 50 states in United States returned questionnaires. Prior to combining samples from different countries, two tests were computed to investigate the potential impact of country affiliation upon the results. Mean comparisons of the interval scaled variables were computed by country affiliation. Using t-tests, no significant mean differences were found on the scores based upon country affiliation. Secondly, data from Canada and all other countries were combined and compared with the United States sample. Again using t-tests, no significant mean differences were found. The results of these tests were the basis for the decision to combine samples.
The respondents were predominantly Caucasian (n = 144, 91%). One hundred and twenty participants were male (76%) and 38 were female (24%). The mean age of the sample was 36 years. Seventy-three percent of the respondents were quadriplegic and 27% were Paraplegic, with an average of 12 years since the time of the injury. Approximately 46% of the respondents had never married while 35% were currently married. Nineteen percent reported being separated, divorced or widowed. Thirteen percent of the sample had earned a college degree prior to their injury, whereas 46% had completed their college education since the time of injury. Seventy-eight persons (49%) were unemployed, 19 persons (12%) indicated working less than 20 hours per week, and 59 persons (37%) were working 20 hours or more each week.
Data Collection Procedures
Using the four formats (WWW, print, telephone interviews, e-text), persons with a spinal cord injury were invited to participate in this study. The invitation letter described the purpose of the study, inclusion criteria, questionnaire formats available, self-disclosure of the researcher's personal Disability, toll-free phone number, and home page address. The letter was then sent via e-mail to five listservs related to disability, posted on a WWW bulletin board, and mailed to organizations focusing on spinal cord injuries and disabilities. The questionnaire that was posted on the WWW was available for six weeks. During that time, requests for participants in this study were e-mailed to various listservs on three occasions. To be as inclusive as possible, the letter requesting participation in this study was also mailed to all 10 Centers for Independent Living in the state of Florida and to 11 model Spinal Cord Injury Rehabilitation Centers throughout the United States.
Using the World Wide Web is quite appropriate for researchers seeking to target a specific population, such as persons with a spinal cord injury (Schmidt, 1997). The WWW allows any person with access to the Internet, regardless of severity of disability, the opportunity to independently and anonymously complete the questionnaire. The method employed for this study simplified the process for participants, particularly persons with Quadriplegia, by removing barriers such as handling multiple pages of paper and the inconvenience of return mail. We have no data as to why most participants chose to complete the WWW form. Perhaps people chose to participate using the WWW format because they can view and answer the questions independently, anonymously, and at their convenience, without the hassle of shuffling papers, which is often difficult for persons with higher level spinal cord injuries.
Instrumentation
Dependent variable. Consistent with the literature review, life satisfaction was used as the dependent variable. Life satisfaction is described as a psychological state that may be broadly associated with psychological well-being (Neugarten et al., 1961). This study used the 18 item Life Satisfaction Index-A (Neugarten et al., 1961) to measure life satisfaction, with a higher score indicating greater life satisfaction. The Life Satisfaction Index-A consists of 18 statements to which respondents indicate whether they agree, disagree, or are undecided. A score for the Life Satisfaction scale is calculated by assigning one point for each appropriate response and summing the values for each participant's score. Scores range from 0-18, with a higher score indicating greater life satisfaction.
Correlations have been demonstrated between the Life Satisfaction Index-A and other measures of life satisfaction: Life Satisfaction Index A (.989), Life Satisfaction Index-Z (.952), Philadelphia Geriatric Center Morale Scale (.779); (Lohmann, 1977 as cited in Decker & Schulz, 1985). While Reliability analysis by Schulz and Decker (1985) resulted in a Cronbach's alpha of .76 for the total scale, it resulted in considerably lower values for individual subscales.
Independent variables. Seven independent variables were examined in this study. Possible influences considered were: perceived control, communication skills, personal assistance satisfaction, marital status, handicap, years since injury, and health.
The first independent variable, perceived control, refers to the belief that a person can direct one's life, has the power to make decisions based upon individual preferences, and can change the course of a person's life (Nosek, Fuhrer, & Howland, 1992). The Perceived Control subscale of the Personal Independence Profile (Nosek & Fuhrer, 1992; Nosek et al., 1992) was used to measure perceived control. A higher score signified more control.
The second independent variable, communication skills, refers to the clarity and directness used in verbal exchanges of information (Epstein, Baldwin, & Bishop, 1983). The Family Assessment Device was originally designed to assess familial interactions. Items in the Communication Subscale such as "people come right out and say things instead of hinting at them" can be applied to various interpersonal relationships. The Communication Subscale of the Family Assessment Device (Epstein, Baldwin, & Bishop, 1983) was employed in this study to measure verbal communication. A lower score indicated greater clarity and directness in the exchange of information.
The third independent variable, personal assistance satisfaction, is ranked the highest priority by persons with disabilities and professionals researching disabilities (English, 1994). Personal assistants serve to maintain the well-being, personal appearance, comfort, safety and interactions within the community, and society as a whole: this includes tasks which individuals would generally do for themselves if they did not have a disability (Litvak, 1991).
The Personal Assistance Satisfaction Index (Nosek, Quan, & Potter, 1992) was used to measure satisfaction with personal care assistance services. A higher score suggested greater satisfaction. Of our respondents, approximately 30% of the sample did not complete the Personal Assistance Satisfaction Index. A cross tabulation revealed that persons with a missing score tended to be those who were married (n = 28) or persons with an injury level (e.g. C5-8 quadriplegia incomplete and Paraplegia) that may have allowed for greater independence (n = 23).
The fourth independent variable, marital status, was coded from the demographic information provided by each participant. Choices included: never married, married, separated, divorced, and widowed.
