Volume 105(4), April 2005 165 The Principles and Standards for School Math- ematics (National Council of Teachers of Mathematics [NCTM], 2000) and the Benchmarks for Science Literacy (American Association for the Advancement of Science [AAAS], 1993) both outline standards for mathematics and science learning. To demonstrate proficiency in data analysis and probability in grades 6 through 8, students should be able to (a) select, create, and use appropriate graphical representations of data, (b) find, use, and interpret measures of center and spread, and (c) discuss and understand the correspon- dence between data sets and their graphical represen- tations, as stated in the Principles and Standards for School Mathematics. The Benchmarks for Science Literacy stated that students in grades 3 through 5 should be able to make graphs, spread out data, and compare two groups of data. The Benchmarks for Science Literacy for grade 6 through 8 stated that students should have many experiences in making data tables and graphs, as well as using them to describe a variety of patterns and relationships. The general inclu- sion of data representation and interpretation within each of these distinct documents is a measure of its importance; therefore, collecting, representing, and interpreting data are important goals to include within the middle grade curriculum. The purposes for graphing lie in the conveyance of numerical data in a visual format (Arvin & Colton, 1940; Biehler, 1989; Cleveland, 1993) and in conveying to the reader the patterns and/or irregularities present in the data that may not be evident in table form (Isaacs & Kelso, 1996). Many students consider the data only as a group of numbers on which they will perform algorith- mic functions, often mitigating graphical representa- tions as an aid to data analysis and interpretation (Konold, Higgins, & Russell, 2000). One of the most important decisions for students to make in the con- struction of a graph is determining which visual method should be used (Friel, Curcio, & Bright, 2001) to answer the question presented. The problems with learning data analysis concepts have been well documented (Fast, 1997; Kahneman & Tversky, 1972; Lynch, Coley, & Medin, 2000; Shaughnessy, 1992). Misconceptions about centrality and data interpretation have been documented in el- ementary school students (Jones, Langrall, Thorton, & Mongill, 1999). At the middle grades, misconceptions tend to focus more broadly on interpretation and select- ing appropriate representations of the data. Secondary school misconceptions have been more prevalent in measures of central tendency and its interpretation. The persistence of these misconceptions is evidenced in undergraduate statistics classes (Capraro, Kulm, Hammer, & Capraro, 2002; Conners, McCown & Roskos-Ewoldsen, 1998). In the present study, the ideas related to producing graphs from data and using these representations to make conclusions and inter- pretations of the data are specifically addressed. Purpose This paper explores some statistical conceptions and misconceptions of middle grades students and the implications for representing and interpreting data. Why do these misconceptions persist and what can be done about them? Hawkins and Kapadia (1984) blamed contemporary education because it forces determinis- tic cognitive strategies rather than supporting the devel- opment of concepts in an indeterministic environment. Cox and Mouw (1992) stated that these misconceptions yield correct predictions sometimes, which makes re- placing them even more difficult. Fischbein and Gazit Middle Grades: Misconceptions in Statistical Thinking Mary Margaret Capraro, Gerald Kulm, and Robert M. Capraro Texas A&M University A sample of 134 sixth-grade students who were using the Connected Mathematics curriculum were administered an open-ended item entitled, Vet Club (Balanced Assessment, 2000). This paper explores the role of misconceptions and naïve conceptions in the acquisition of statistical thinking for middle grades students. Students exhibited misconceptions and naïve conceptions regarding representing data graphically, interpreting the meaning of typicality, and plotting 0 above the x- axis.