How to Design and Evaluate Research in Education

CHAPTER 23 Mixed-Methods Research 571researcher discussed the project with administrators and other staff to obtain initial approvalto solicit participants and provided all the materials for the schools and organizations.In some cases, the primary researcher went to the sites to administer the surveysonce parental consent forms were obtained, and in other cases, the site staff administeredthe surveys and returned them, along with the consent forms, to the researcher viamail. Several sites met with large groups of students on a regular basis (e.g., TRIOprograms) and thus were able to monitor the return of informed consents in order toensure maximum participation by students. For these sites, in addition to those fromschools and community programs, the response rate was approximately 65%. At oneafterschool program, however, 100 consent forms were given to supervisors to pass outto students, and only 7 were returned. This was surprising considering the relatively highresponse rate we obtained from other sites, and we are unclear as to what extent thestudy was actually described to students as was intended at this particular program.Packets containing two informed consent forms (one for the students/parents tokeep and one to return to the investigator), as well as an introductory letter, weredistributed to students to take home during school or during their program’s activitytime. Both consent forms and the letter were translated into Spanish such that all parentsreceived copies in English and Spanish. Parents were asked to send signed consentforms back to school (or the organization) with their children. Once consent was obtainedfrom parents, students who volunteered to participate in the study were asked tocomplete a student assent form and then were administered a packet of materials duringa 45-min period of school or of an afterschool program. Only students whose parentshad provided consent and who had themselves completed an assent form were allowedto complete the packet of questionnaires, and the questionnaires were only provided inEnglish. This decision to only sample students who were proficient readers in English wasmade during the development of the project because there was no existing data regardingthe conceptual and functional equivalence of several of the measures for Latinoadolescents in particular (American Psychological Association, 2002; Rogler, 1999).Limits generalizingInstrumentsDemographic questionnaire. A demographic questionnaire was included to obtainparticipants’ age, year in school, gender, race/ethnicity, generational status, and religiousaffiliation.GivensimultaneouslyOpen-ended question about well-being. At the bottom of the first page (demographicform) of each packet of measures was the following open-ended question: “Whatfactors do you think contribute to life satisfaction and happiness?” Students were providedwith 10 lines on which to write their responses and were encouraged to write onthe back of the page if they needed more space.QualitativePerceived social support. The Multidimensional Scale of Perceived Social Support(MSPSS; Zimet, Dahlem, Zimet, & Farley, 1988) is a 12-item scale that measures perceivedsupport from three domains: Family, Friends, and a Significant Other. Participantscompleting the MSPSS are asked to indicate their agreement with items on a 7-pointLikert scale ranging from 1 (very strongly disagree) to 7 (very strongly agree). A sampleitem from the Family subscale is “I get the emotional help and support I need from myfamily.” Support for the reliability and validity of the MSPSS has been found withsamples of college students, adolescents living abroad, and adolescents on an inpatientpsychiatry unit (Canty-Mitchell & Zimet, 2000).