|
Article Excerpt
This study describes the development of the Mathematics Interest Inventory. Internal reliability and concurrent and construct validity were evaluated using 3 samples of children totaling 724. Exploratory and confirmatory factor analyses revealed the presence of 3 factors representing the psychological state of individual interest. Cross-validation with older children supported the factors.
**********
Today's society is technological in nature, resulting in an increasing demand that citizens possess an adequate understanding of mathematics. It is interesting that many continue to find the application of mathematical knowledge at even basic levels to be quite difficult (Bruning, Schraw, & Ronning, 1999), leaving educators with questions about how to remedy this problem. Because students' interest in academics steadily declines as students advance through the school years (Hidi & Harackiewicz, 2000), possibly starting as early as the third grade (Murphy & Alexander, 2000), an understanding of mathematics interest may be an important component in addressing mathematics concerns.
A renewal in interest research has emerged since the decline of the influence of behaviorism and increased emphasis on the study of only observed variables (Schraw, Flowerday, & Lehman, 2001). Interest influences learning through its direction of our attention (Hidi, 1990), its encouragement of specific learning strategies (Wade, Schraw, Buxton, & Hayes, 1993), and its effect on our emotional engagement (Schiefele, 1999; Schraw, 1998). Because of the integral role of interest in learning, the ability to measure this construct effectively is of particular importance not only to continued research but also to practical situations. Educators who can measure students' levels of interest in mathematics will gain valuable information concerning how to educationally program for an individual student and what interventions might be necessary to encourage the student's interest. Mitchell (1993) noted that a major weakness of interest research was the small number of interest measures available. Although this weakness has been addressed to some degree, a continued need exists to develop measures that assist in evaluating interest across specific domains. The purpose of the present study was to develop a measure of mathematics interest and to evaluate the psychometric qualities of the resulting scores.
Although interest has been defined and described in a number of different ways, such as an aspect of personality, a type of motivation, and a result of the individual characteristics of various tasks (Hidi, 2000; Renninger, 2000), most researchers currently agree that two types of interest exist, situational and individual (Hidi, 2001). Situational interest results from an attention-holding, often novel situation or event that offers an ideal level of challenge (Chen, Darst, & Pangrazi, 2001; Hidi, 2001). This type of interest might be associated with instant enjoyment, as Chen et al. (2001) suggested, or in some cases associated with negative emotions as Hidi and Harackiewicz (2000) described with their example of medical students finding the dissection of cadavers to be interesting but not likable. Regardless of the type of emotion elicited, situational interest typically fades as the event or situation changes. However, over time, this situational interest could develop into individual interest (Alexander, 1997; Hidi & Berndorff, 1998), which refers to a relatively enduring disposition associated with an increase in one's knowledge, value, and positive emotion in a specific domain (Hidi & Harackiewicz, 2000).
Many researchers have emphasized the study of the interestingness of situations, because environments might be more easily changed to interest students than working to build an enduring disposition in students toward a particular academic domain (Chen et al., 2001). Even so, the ability to measure individual interest continues to be valuable. The purpose of developing curriculum that encourages situational interest is to facilitate learning but also, in many cases, to facilitate individual interest. The ability to measure individual interest is imperative if researchers intend to identify when and how such stable dispositions develop. In addition, the measurement of individual interest has a practical application. Understanding what levels of individual interest are brought to the classroom by individual students informs educators what types of learning strategies and degrees of persistence their students will likely put forth, enabling them to better plan for interventions and learning activities. This type of measurement is more complex than the level of interest often assessed by typical career interest inventories. A measure of individual interest relevant for the educational setting must first be domain specific, taking into account the interaction between the individual and his or her environment (Krapp, 1999), and second consider the presence of three distinct qualities of individual interest: knowledge, value, and positive emotion (Hidi & Harackiewicz, 2000). Finally, a measure of individual interest for a specific domain, mathematics in the present case, must be associated with reliable and valid scores. The goal of the present study was to assess the aforementioned standards through the evaluation of internal consistency, correlational analyses, and factor analysis of a newly developed inventory, the Mathematics Interest Inventory (MII), based on the discussed definition of individual and situational interest.
METHOD
Participants
Three independent samples with an increasingly older average age were collected to evaluate the MII. Children of increasingly older age were selected so that the developmental appropriateness of scores associated with the MII could be investigated. The first sample comprised fourth and fifth graders (N = 170) attending public schools in West, East, and South Texas. Students' ages ranged from 9 years to 12 years (M = 10.13, SD = .80). Forty percent of the students described their ethnicity as Hispanic, 1% as Black, 40.5% as White, 2% as Native American, .5% as Asian, and 8% as other; 8% failed to respond. The sample included slightly more boys, with 55% of students identifying themselves as boys and 45% as girls.
The second sample comprised fourth and eighth graders (N = 282) attending public schools in West Texas. Students' ages ranged from 9 years to 15 years (M = 11.64, SD = 2.00). Twenty-six percent of the students described themselves as Mexican American, 1% as African American, 1% as Asian American, 3% as Native American, and 68% as White. Three students failed to respond. The second sample included slightly more girls, with 47% of the students identifying themselves as boys and 53% as girls.
The third sample comprised 8th, 9th, and 10th graders (N = 272) attending public school in West Texas. Students' ages ranged from 12 years to 18 years (M = 14.63, SD = 1.22). Twenty-three percent of the students described themselves as Mexican American, 2% as African American, < 1% as Asian American, 2% as Native American, and 69% as White. Eight students failed to respond. Finally, the third sample was evenly split between boys and girls.
Instruments and...
|
