By Clifford Eberhardt, Ed.D
In a recent conversation with one of my graduate students, who is also a teacher in the public school system, we discussed a student’s achievement problem she was having with one of her students, and I ended up giving her advice on a problem. She had concerns about why a certain student of hers was scoring very low on tests, when she knew he knew the material? “Try Triangulation!” I suggested. In others words, try comparing other educational and social variables in the student’s sphere of learning that may have an impact on the low scores.
Triangulation is a research tool used to compare unlike data sets of a particular person, place or things to determine if relationships exit between different variables under analysis. Triangulation in educational research is a form of cross comparison of unlike data and seeks to find patterns of student behavior, negative social handicaps facing students, irregular study habits and other variables that may impair student achievement. The goal is to collect data from different participants and settings using comparative research methods to identify recurring results.
In collecting data for triangulation to solve problems in education like the case above with my graduate student, you should know that most of the data used in this triangulation model are readily available. In most cases, students’ test scores, students’ absentee records, students’ missed assignments and other variables about students already exist and are just a mouse click away on your computer.
For example, test scores can provide some information about a particular student’s behavior on a test, but for more in-depth information about the student, you can triangulate test scores with: a. the number of student absentees, b. the number of student missed assignments, c. the parental profile (single parent or not) of the student, etc. By triangulating test scores with other educational and social variables, you can gain new knowledge about the test scores of a student.
The benefits accrued through the use of triangulation as both a design strategy and an analytic tool cannot be overstated (Schwartz, Terry Ann; Kaplan, Michael H., 1981). Triangulation allows for the clustering and organizing of disparate yet related data. Finding out what the data have in common and how the data are different allow the researcher to eliminate… or reduce the number of plausible rival hypotheses that might explain such data.
Thomas and Znanieck, in their investigative report, “An example of Multiple triangulation: The Police Peasant in Europe,” found that police departments in both Europe and America use triangulation to solve crimes.
In their investigation, Thomas and Znanieck, found five basic types of triangulation:
1. Data triangulation, involving time, space, and persons,
2. Investigator triangulation, which consist of the use of multiple, rather than single observers,
3. Theory triangulation, which consists of using more than one theoretical scheme in the interpretation of the data,
4. Methodological triangulation, which involves using more than one method and may consist of within-method or between-method strategies,
5. Multiple triangulation, when the researcher combines in one investigation multiple observers, theoretical perspectives, sources of data, and methodologies.
Retrieved from the Internet on, 8/4/07 —
Consciously or subconsciously, we triangulate problems and situations everyday. Anytime we have a problem to solve or information to add, we triangulate. For example, when a person is considered for a job, their prospective employer starts multiple triangulations on them from the information in their vita and job application. The education that they proclaim to have on their application is triangulated back to the university they went to. Their job experience stated is triangulated to the different past jobs they listed on their job application. The opportunities for triangulation are endless when you consider that we are a society obsessed with data gathering and collection.
Greta Morine-Dershimer, in her paper published by the “Communication Services, Research and Development Center for Teacher Education, Education Annex 3.203 — The University of Texas at Austin,” stresses that research should be at the heart of triangulation. In her paper, “Tapping Teacher Thinking through Triangulation of Data Sets,” Morine-Dershimer examines how “Triangulation,” is used in several data sets to test, corroborate and elaborate on data, and is one method for increasing validity of findings. Her paper illustrates a process of triangulation of data obtained by three different data collection procedures: stimulated recall interviews; Kelly Repertory Grid interviews; and ethnographic observations of classroom interaction (Morine-Dershimer, G. 1983).
Beth L. Chance and Joan B. Garfield in their paper, “New Approaches to Gathering Data on Student Learning for Research in Statistics Education,” while stating little about triangulation per se, their methods for gathering data in the classroom is classic triangulation. In their paper they did a critique on the most effective approach of gathering research data on two methods of teaching statistics; classroom-based research or videotaped interviews/observations of students. Both methods are reliable for gathering data in the classroom, but with triangulation a statistic’s instructor can determine the most effective method (Chance and Garfield, 2001).
Data analysis is one of the cornerstones of triangulation. The goal is to collect and compare data from different independent sources. Denzin identifies three analytical methods used in triangulation: 1 comparing multiple sources of data across participants, time and sites; 2. comparing the results of multiple independent investigations; 3 comparing multiple methods of data analysis. Other approaches to triangulation include observing participants’ consistency in different situations or comparing participants’ consistency in their verbal statements and actual performance. Either way, the ability to produce reliable results from different times or methods enhances the credibility of the data (Denzin, 1978).
Even though triangulation can be a useful research tool, it still has to abide the same rigor as the scientific method for the results to be valid and reliable.
Steps in creating a triangulation model using the scientific method
Ask questions to identify a problem: The scientific method starts when you asking a question about something that you have observed: a problem. A problem always stems from a need. A need is the discripincy between the present state of things and the desired state of things. In the problem with my graduate student, the state of things was that a certain student of hers was scoring low on tests, when she knew he knew the material… the problem. The desired state of things is for her to find out the problem behind the low scores: the solution. However, in order for the scientific method to solve the problem, it must be about something that can be measured, preferably with a number.
Do background research on the problem: Rather than recreating the wheel in putting together a plan for solving your problem, you want to be a smart researcher by using library resources and Internet research to help you find the best way to address the problem. This way, you don’t repeat mistakes made by others the past. Moreover, you want to interview other teachers who may have had the same problem. You may want to interview students who may have had the same problem and others who can be of help in understanding the problem.
Construct a hypothesis of the problem and collecting data: A hypothesis is an educated guess about what the problem may be, and of course, your hypothesis should be constructed in a way to help you address the original need: to find out why a student is scoring low on tests, when he knows the material. Collecting the data for this problem is easy because these data already exist; the student’s absentees, number of missed assignments, parental profile… all of which are independent variables in this problems and are at your access.
Analyze your data and drawing a conclusion about the problem: Once you have triangulate low tests scores with he student’s absentees, number of missed assignments, parental profile, us an except or reject your hypothesis. Each triangulation or hypothesis will produce a true or false conclusion. In many triangulation attempts, the researcher will construct a new hypothesis starting the entire process of the scientific method over again. Even if they find that their triangulation or hypothesis was true, they may want to test it again in a new way.
The Results: The results of your triangulation can be used for action on solving the problem; for further research on the problem or can shared with other educators who have experienced the same problem… via journals or over the Internet.
While triangulation is not a silver bullet for solving problems in the classroom; it didn’t even work for my graduate student, her student had some can of test anxiety that was more psychological then pedagogical. With that being said, today’s educators will need more the books, erasers and chalkboards to solve the educational problems in of the future. Triangulation is a good start!
Clifford Eberhard is a graduate professor at Central Michigan University.
1. An example of multiple triangulations: The Police Peasant in Europe and America by Thomas and Znanieck’s. Their investigation uses triangulated data
Retrieved from the Internet on, 8/4/07 —
2. Beth L. Chance and Joan B. Garfield in their paper, “New Approaches to Gathering Data on Student Learning for Research in Statistics Education.
3. Denzin,N (1978) The research act. Newbury Park, CA: Sage
4. ERIC #: ED202904
Title: Operationalizing Triangulation in Naturalistic Evaluation: Community Education in Kanawha County (WVA).
Authors: Schwartz, Terry Ann; Kaplan, Michael H.
Descriptors: Case Studies; Community Education; Data Collection; Evaluation Methods; Program Evaluation
Publication Date: 1981-04-10
5. R&D Rep. No. 8014.
Title: Tapping Teacher Thinking through Triangulation of Data Sets
Author: Morine-Dershimer, Greta
Descriptors: Classroom Observation Techniques; Classroom Research; Classroom Techniques; Cognitive Processes; Comparative Analysis; Data Analysis; Interaction Process Analysis; Junior High Schools; Research Methodology; Secondary School Teachers; Student Behavior; Student Characteristics; Teacher Response; Teacher Student Relationship; Teaching Methods
Publisher: Communication Services, Research and Development Center for Teacher Education, Education Annex 3.203, The University of Texas at Austin, Austin TX 78712.
Publication Date: 1983-06-00