FAQs AP Research Surveys Publications
Frequently Asked Questions about the FICSS Study

Q: What is the goal of FICSS?
A: FICSS is designed to identify measurable factors in high school science classes that predict different levels of achievement in introductory college science courses.

Q: How does FICSS gather data?
A: The core data comes from a detailed six-page survey completed by college students early in their introductory college science course, together with each student’s final course grade supplied by the professor.  Professors remove any personal identifiers before sending us the surveys, insuring individual confidentiality for each student.  Each survey is identified only by a unique record number.

Q: Who has participated in FICSS?
A: Participants include nearly 18,000 students and their professors at randomly selected colleges and universities across the United States; about 2,000 additional randomly chosen college professors; about 1,000 of the students’ previous high school teachers; and another 3,000 or so randomly chosen high school teachers.

Q: What subjects do your participants teach or study?
A: The participants teach or study Physics, Chemistry, or Biology.

Q: Did you do any research on teachers and professors?
A: Part of the FICSS project was a Correspondence Survey which compared the ideas, beliefs, and assumptions of college professors and high school teachers.  Our previous interviews with members of both groups suggested a sharp division about what they thought was important to prepare students for college science.  The Correspondence Survey is designed to investigate such differences, as well as similarities.

Q: Have you included questions about the attitudes and feelings about science developed by students in high school?
A: We included very few direct questions about affective or emotional factors for the following reasons.  Our experiences (and previous research literature) indicate that student self-reports about previous emotional states are very unreliable.  Self reports about concrete experiences, however, have been shown to be quite reliable, even four years after the fact.  On our survey we asked about frequency of class meetings and labs, size of class, types and frequency of exams, typical homework load, topics covered, etc.

Q: But emotional aspects are crucial to learning science.  Can you say anything about this from your study?
A: Yes, in three ways.  First, we included questions about factors that are known (from research) to contribute to positive or negative attitudes about science.  Secondly, we asked about detailed personal/demographic factors that are thought to contribute to attitude differences (science hobbies, parental science encouragement, etc.).  Thirdly, we allowed students to voluntarily supply their e-mail addresses on the survey, then administered an on-line survey about their high school science experiences designed to tap into emotional aspects.  Many of the on-line questions were open-ended short answer opportunities.

Q: What has FICSS proven about how high school affects college learning?
A: Nothing.  Our research design was patterned after epidemiological studies, so our findings are statistical in nature.  Just as epidemiological researchers can identify factors – such as smoking – that can predict lung cancer, we are identifying high school factors that can predict differential performance in introductory college science courses.  Our findings are based on analyses by multiple statistical methods shown to be effective in large, randomly selected samples such as ours.  We will only publish results that meet strict scientific standards of validity and reliability.  As in the proposed smoking/cancer link, we will identify factors to be further investigated in enough depth to identify causality.

Q: What assumptions guided your research design?
A: We made two assumptions at the start of our investigation:
1.  Students’ high school science experiences can affect their college science performance.
2.  It is possible to investigate the high school/college science link with an epidemiological-style study.
Beyond these assumptions, we designed our research with as much objectivity as is humanly possible.  In fact, some of our initial findings challenge many of our own assumptions.