One of the lightning rods in the discourse over teacher pay has been the question of “how much” teachers should be paid. What the “how much” debate does not directly address, however, is the question of “how” teachers should be paid.

This paper attempts to help lay groundwork for a better understanding of what exactly teachers want in terms of how they should be paid.

Nontraditional teacher pay (NTTP), which is considered here to be pay systems that diverge from exclusive reliance on the traditional salary schedule, continues to be a high-profile and controversial issue. Often lost in the debate over NTTP is a thorough assessment of what teachers actually prefer in a pay system, which has important implications for teacher recruitment, retention, and motivation. While teachers are sometimes broadly described as resisting NTTP, the truth is much more nuanced. Using data collected in 2005 from over 2,500 unionized teachers in a single state, this paper explores: (1) whether teachers favor or oppose four different bonus-based NTTP systems, and the demographic and attitudinal characteristics associated with these preferences; (2) levels of teacher support for traditional (i.e., education, service) and nontraditional (e.g., standardized test score) criteria for salary increases, as well as teacher characteristics related to this support; and (3) changes over time in NTTP preferences, as additional survey data from six years earlier allow for a unique opportunity to examine possible cohort versus service effects.

The main findings are:

–Teachers supported an emphasis  on education and service as the basis for salary increases, preferring these criteria to student test scores and performance evaluation.

–Teachers with more service  and education preferred service and education, respectively, as salary increase criteria. Moreover, possessing more service also appeared to cast doubt on the suitability of all alternative salary increase criteria, predicting lower support for education, performance evaluation, and test scores as criteria.

–Merit-based bonus plans were the least favored of the four bonus plans presented, as only 28% of teachers favored their adoption. Teachers opposed the merit bonus plan even though there was no downside risk associated with the potential add-on pay (possible explanations include teachers suspecting that money for the bonus pool will ultimately mean less money for the salary schedule increments, the violation of teachers’ strongly held norms of equity, and an aversion to any perceived subjectivity in pay allocations that might make favoritism more likely).

–Agreement with whether a teacher’s performance evaluation should be an important factor in salary increases fell about midway between the embracing of service and education as criteria and the rejection of test scores.

–Thirty-five percent of the teachers here either agreed or strongly agreed that job performance should play an important role in salary increases. When asked to report how much of a role, however, many of these merit pay “supporters” retreated from this position, as half of them indicated that only one-fifth or less of their pay increases should be linked to performance evaluation (in fact, 16% of this supposedly pro-merit group actually recommend that 0% of pay should be tied to performance). Assuming a generous 5% merit increase pool, this means that half of the teachers that supported the use of merit-based salary increases advocated having only 1% or less of their pay dispersed via a merit increase format.

–Fifty-two percent of all teachers recommended that 0% of salary increases be tied to performance evaluation, while another 25% of teachers recommended that performance evaluation be used to determine just 1-10% or 11-20% of the increase.

–Teachers were strongly opposed to the use of student test scores as bases for salary increases. While 83% of teachers either disagreed or strongly disagreed that the school’s student test scores should be an important factor in their pay increases, 87% of teachers either disagreed or strongly disagreed that their own students’ test scores should play such a role. Teachers also rated the linking of pay to standardized test scores as the most ineffective of several potential methods for improving academic performance.

–Job satisfaction and the belief that the school administration acts fairly in dealing with teachers (procedural justice) reduced aversion to test scores as a pay determinant. However, even at very high levels of these two attitudes, teachers still clearly disagreed with such use of test scores. Thus, despite high procedural justice perceptions and high job satisfaction alleviating test score concerns to some degree, the opposition remained unambiguous and robust.

–Samples from both 1999 and 2005 provided an opportunity to determine whether the tendency for teachers with more service to oppose NTTP was the result of (a) a cohort effect (recent cohorts being more NTTP-friendly upon entering teaching than were earlier cohorts), or (b) changes within teachers as they accumulate service. There was no support for a cohort effect. Evidence indicated that the average teacher becomes less NTTP-friendly over time, growing less likely to favor the adoption of the bonus-based NTTP plans or to support the use of NTTP criteria in salary increases.