WWW Site for John Lawrence Bencze, Associate Professor, Science Education, OISE/University of Toronto

Procedural Education
Developing Expertise for
'Judging Ideas'

Developing Expertise for
Correlational Studies
Welcome! This page provides perspectives, general practices and links to resources for helping students to develop expertise for conducting studies in science & technology. Correlational studies can be used for Judging Ideas, which is a major phase of my constructivism-informed pedagogical framework. If you have comments, questions, suggestions, resource ideas, etc. about anything here, please write to me about them. Thanks.


NoST/STSE Regarding Studies
When scientists & technologists seek empirical support for claims about phenomena, they often use studies. A study is an approach that should be distinguished from 'experiments,' about which perspectives and practices are provided elsewhere. As indicated in the figure here, a study involves collecting data regarding variables that change naturally. To develop ideas about food decay, for example, we could find some food-bacteria mixtures in nature, measure their temperatures and determine the rate of decay in each case. This particular inquiry would be a correlational study, because investigators would be comparing (correlating) how changes in one variable (e.g., natural temperature changes) may affect changes in another variable (e.g., rate of decay). Again, studies should be contrasted with experiments, which involve forcing changes to occur in a possible 'cause' (independent) variable. In the figure here, a decay experiment would involve steadily increasing the temperature (an independent variable) at which food-bacteria mixtures are kept and measuring the rate of decay of the food (a dependent ['result'] variable).

Which of these two sorts of investigations inquirers choose to use depends on various factors. Experiments often produce quick results and, for that reason, they may be preferred over studies. Sometimes, however, experimentation is impractical and, therefore, studies are necessary
— as in the case of much of astronomical studies, for example. Perhaps more significantly, it often is ethically problematic to conduct experiments in some cases. With every experiment, there is a risk of negative side-effects of forcing a variable to change. Of the numerous examples, asking volunteers to smoke different numbers of cigarettes in order to deterine cigarettes' effects on human heart rate may work, but it risks giving them lung cancer!
— along with other possible unexpected negative side-effects.

Strategies for Helping Students to Develop Expertise for
Conducting Correlational Studies

There are various approaches for helping students to develop skills they could use for conducting correlational studies, but resources here are based on the constructivism-based approach at Skills Pedagogy. Resources for the Modelling & Practice stages of this framework are provided on this page, through the links at right. Resources for helping students to express their pre-instructional skills for science inquiry (including experimentation) are provided at Expressing Inquiry Skills.

Teacher Modelling
(including Teacher Demonstrations, Student Reading)

Interactive Teacher Demonstrations: Teachers can engage students in Socratic discussions about studies, first pointing out differences between studies and experiments perhaps by discussing what would be controversial studies; such as a marijuana study of rates of human coughing (which some students may realize also might cause cancer). More extreme inappropriate uses of studies also could be mentioned, such as those performed by the Nazis during the Holocaust. More benign examples also may be used, like these. A demonstration like that below could then be used to reinforce these ideas. During and after such discussions, teachers should ensure students at least begin to develop understandings of important principles relating to studies — including, for example, those ideas described briefly at right.
Characteristics of Studies
  • a study involves collecting data relating to natural changes in one or more variables, some of which may be causally related (e.g., temperature changes and rates of food decay);
  • relatively large sample sizes should be studied, such that: i) a reasonably wide range of values of the cause variable (e.g., 0 - 90 oC) is collected, ii) there is some 'saturation' of data items within the range of values of the cause variable (e.g., measurements at relatively regular intervals, such as 0, 10 , 20, 30 oC), iii) there is duplication and/or near duplication of data items at relatively regular intervals (e.g., two or more temperatures at or near each of 0, 10 , 20, 30 oC);
  • like experiments, studies should be 'controlled,' but in a different way; e.g., by: i) selective sampling (e.g., by studying decay in roughly the same kind of soil), ii) ignoring data that appear to be caused by unusual factors (e.g., studies of decay in areas receiving excessive rainfall);
  • where feasible, quantitative and qualitative observations of cause and result (dependent) variables should be made;
  • where feasible, related, comparable result variables (e.g., leaf colour & thickness) should be studied;
  • studies should be used in conjunction with theory-development; e.g., data from studies may not mean much or be misleading without appropriate theory.
  • studies are particularly useful for evaluating effects of potentially harmful substances and processes, leading to what can be recognized as WISE Problems.
A common demonstration, which can later be used for student practice, involves use of a ruler (of varying sorts) to measure a person's reaction rate. The "0" end of a ruler is held just above the imaginary plane from the top of the person's thumb to the top of his/her forefinger, as indicated at right. The person should hold his/her thumb and forefinger at a particular distance (e.g., the width of the ruler). The person holding the ruler should pretend to drop the ruler a few times to ensure the subject (person attempting to catch the ruler) does not react too quickly and unfairly. The person holding the ruler should drop it without indicating it is going to be dropped. Assuming the subject catches the ruler, the point at which it is caught (e.g., at the top of the person's finger and thumb) should be recorded. This distance measurement can be converted into a sort of 'rate' measurement by dividing 1.0 by the number. This produces a ratio that would increase as a person gets faster at reacting (i.e., a short distance is taken to catch the ruler). The person can be given three trials, perhaps, and an average calculated. Having demonstrated how this measurement works, the teacher could ask students to brainstorm possible causes for why people might have different reaction rates. Some possibilities include: amount of sleep, length of fingers, related hobbies (e.g., piano playing), sex (male vs female), age, distance between the eyes, etc., etc. A series of students may volunteer to be studied during the demonstration, but this study likely is best conducted as a whole class practice activity — as discussed below. In discussions about possible cause variables that could affect reaction time, you could get students to consider variables like cigarette-smoking and alcohol consumption - as connections to WISE Problems.
There are several other demonstrations involving observations about students that teachers can use for providing instruction in the nature of studies. Some other variables that could be measured and then correlated with other variables are: i) various body dimensions (e.g., arm length and leg length, heart rate and body weight, etc.).

Again, in conducting these demonstrations, teachers should be careful to ensure students understand points like those above about studies.
Student Reading: Teachers should encourage students to review various sources of information about the nature of studies and, especially, correlational studies. Some reading of this sort is provided with some of the practice activities available below. Information also is available on various Web site, such as at Correlational Research and through Skills Ed. Resources.

Student Practice
Students should be given several opportunities to gain some practice in developing expertise regarding studies. Generally, these may involve: i) designing & conducting studies and/or ii) analyzing & evaluating completed inquiries. As students work through the various activities, they should have opportunities to interact with peers. Teachers also should interact with students on a regular basis, questioning them about what they are doing and reasons for their actions. Generally, teacher interactions should encourage students to develop independent experitse, rather than developing a reliance on the teacher for the 'right' answers, methods, etc. Some specific suggestions for these sorts of activities are provided at right, along with links to relevant resources.
  • A pragmatic approach is to provide students with some sets of variables involving phenomena for which equipment and materials are easily available — such as those @ A, B, C (with template for reaction timers, to be copied onto heavier weight paper), D. A related exercise is to have students consider a phenomenon (e.g., tennis ball bouncing) and design an experiment and/or a study, perhaps using a guide sheet like this one.
  • In terms of analysis & evaluation of projects involving studies, and excellent source is students' science textbooks, which have some 'labs' or 'inquiries' that involve studies — although they are sometimes referred to as 'experiments.' Another very good source is newspapers and popular health magazines, which commonly reports methods & results of studies. Many such reports also are available on the Web, including many 'clinical trials' that involve both studies and experiments. For these, teachers may want to develop appropriate questions for student analysis and evaluation.
All rights reserved, J. L. Bencze, 2008.