Evaluation of a software prototype designed to teach landscape cost estimation

NACTA Journal, Mar 2003 by Henry, Paul, Midden, Karen, Mings, Shara

Evaluation of a Software Prototype Designed to

Teach Landscape Cost Estimation1

Abstract

A project was undertaken to develop an interactive software program that could be used to instruct students in the process of landscape cost estimation. The objective of this study was to evaluate an early prototype of this software and compare its effectiveness, as a teaching tool, to the more traditional type of

instruction consisting of lecture and overheads. Pre/post testing indicated that both instructional methods increased student understanding about the subject, and the software proved equally effective to the lecture in this regard. Students generally regarded the software as a more entertaining delivery platform, and thought that the software would be a useful teaching tool when completed.

Introduction

The business of landscape horticulture has become increasingly competitive. It is estimated that only fifteen percent of newly established landscape horticulture companies survive until their tenth anniversary (Ross, 1993). The reasons for this high failure rate are uncertain but one contributing factor is that students preparing to enter the industry usually receive little, if any, academic training in the area of cost estimation.

Cost estimation is defined as the "art of adding markups to costs in order to determine a price for a job" (Hutson, 1994). It is a complex process consisting of many different steps that must be performed in a logical sequence.

The process, while very difficult for students to conceptualize, also proves challenging to horticulture instructors who attempt to teach the subject. Most landscape horticulture programs do not offer full semester courses in cost estimation and time allocated to the subject often consists of a single lecture and perhaps one lab session.

Because of the difficulties mentioned above, computer-based instruction may prove the best method to teach students the subject of cost estimation. Computer-based instruction has been found to have beneficial effects on achievement in a wide variety of instructional settings (Draper, 1996; Roblyer, 1985; Kulik and Kulik, 1987). Benefits include individualized instruction which may be paced to the needs of the student, cost-effectiveness (particularly when compared to outdoor lab exercises), and increased efficiency with respect to the amount of time it takes students to learn a particular subject. In addition, students may be better able to develop critical thinking and problem-solving skills when using computer based instruction (Rhodus and Hoskins, 1995).

A project was initiated at Southern Illinois University Carbondale (SIUC) in 2000 to develop an interactive software program that could be used to instruct students in the process of landscape cost estimation. The objective of this study was to evaluate a prototype of this software and compare its effectiveness, as a teaching tool, to the more traditional type of instruction consisting of lecture and overheads.

Materials and Methods

Fifty-three students enrolled in a landscape design class within the Dept. of Plant, Soil and General Agriculture at SIUC were randomly assigned to either a control group (27 students) or treatment group (26 students). All students took a pretest (Table 1) designed to assess the students' existing knowledge of cost estimation. Students participating in the study had received no formal academic training in cost estimation prior to administration of the pretest.

One week after administration of the pretest, students in the control group attended a lecture presentation on cost estimation in a typical classroom setting using the same room in which the class was normally held. The lecture format included use of overhead projections. Students in the treatment group, who were not exposed to the lecture, were taken to a multimedia center on campus and asked to view the software prototype that had been developed. Both the control and the treatment groups received the same information with respect to cost estimation although the delivery formats were different.

Immediately following the lecture and software presentations, a posttest (Table 1) was given to both groups of students. The same set of twenty questions was used for both the pretest and posttest. All questions were presented in a Likert scale format with possible responses ranging from one (disagree) to five (agree). Control and treatment groups were compared with respect to their mean scores on both pre- and pretest using t-test at the 5% significance level. Then the difference between the pretest and posttest scores was determined for each student.

T-tests were performed to determine whether the difference between pretest and posttest scores were significant for each question within each group of students (Table 2).

Following administration of the posttest, a qualitative survey (not presented) was given to all participants in both the control and treatment groups. The survey was designed to assess student perceptions about the presentation formats, and did not ask questions specifically about cost estimation. Students were also encouraged to make written comments concerning any aspect of the testing procedure.