Cook v. Stewart , 28 F. Supp. 3d 1207 ( 2014 )

ORDER ON CROSS-MOTIONS FOR SUMMARY JUDGMENT

MARK E. WALKER, District Judge.

Before this Court are the parties’ cross-motions for summary judgment, ECF Nos. 84, 86. This Court has already resolved some of the issues addressed in these motions and responses thereto, see ECF No. Ill, but other issues in the cross-motions are unresolved. In particular, Plaintiffs’ substantive due process and equal protection challenges to the evaluation policies adopted by the school boards of Alachua County, Escambia County, and Hernando County — and approved by the State Board of Education and Department of Education — remain viable.¹

I. BACKGROUND²

The background of this case .has already been discussed extensively by this Court in its previous order on a motion to dismiss. See ECF No. 111. Nonetheless, because the districts’ evaluation policies, and not the Student Success Act itself, are now the objects of judicial scrutiny, some background on these policies is necessary.

The Student Success Act (“Act”), passed by the Florida Legislature in 2011, requires school districts to evaluate teachers and other instructional personnel based in part on “student learning .growth” beginning with the 2011-2012 school year. § 1012.34(3)(a), Fla. Stat. The Act mandates that “student learning growth” be “assessed annually by statewide assessments or, for subjects and grade levels not measured by statewide assessments, by school district assessments.... ” Id. For those courses “associated with statewide assessments,” the Act directs school districts to use specific formulae developed by the Commissioner of Education to measure student learning .growth; for all- other courses, school districts are directed to select an “equally appropriate formula ..., except as otherwise provided” in section 1012.34(7), Florida Statutes. Id.

The formula selected by the Commissioner of Education pursuant to 1012.34(7)(a), Florida Statutes is a so-called “value-added model.” ECF No. 84, at 2. Such a model purports to “analyze student ... test scores collected over a period of time with the intent of separating factors unique to students and schools from factors unique to a classroom teacher to attribute growth in student achievement to teachers and schools.” ECF No. 86-2, at 71. The model employed in Florida, known as the FCAT VAM, essentially takes as input student test scores on the FCAT and a host of “predictor variables” — test scores in prior years, the number of subject-relevant courses in which a student is enrolled, attendance — and outputs measures of individual teachers’ contributions to student success (the “teacher effect”) as well as the so-called “school component” of student success, which “captur[es] the latent effect of all potential impacts of the school community, including principal leadership, neighborhood effects, etc.” Id. at 72-76. The “final” teacher evaluation score consists of the “teacher effect” added to one-half of the school component.³ Id. at 73-75. The FCAT YAM was designed to evaluate teachers who teach courses associated with the FCAT and whose students have a baseline score for calculating the VAM, which means it applies to teachers who teach reading in grades 4 through 10 and math in grades 4 through 8 (hereinafter “Type A” teachers or FCAT teachers).

There are'currently a limited number of other courses “associated with statewide assessments,” see ECF No. 84, at 2-3, and school districts are not required to have district assessments in place for all courses until the 2014-2015 school year, see § 1008.22(6), Fla. Stat. The practical result of this has been that many school districts, including Alachua, Escambia, and Hernan-do, have availed themselves of the temporary⁴ provisions of section 1012.34(7)(e), Florida Statutes. ECF No. 86, at 1-3.-This subsection explicitly allows teachers who teach students who take statewide assessments, but who teach them in courses other than courses “associated with” those statewide assessments, to be evaluated based on the test scores of their students:

For classroom teachers of courses for which the district has not implemented appropriate assessments under [section] 1008.22(8) or for which the school district has not adopted an equally appropriate measure of student learning growth ..., student learning growth must be measured by the growth in learning of the classroom teacher’s students on statewide assessments....

Id. § 1012.34(7)(e) (emphasis added). (For the purposes of brevity, these teachers who teach students in grades 4-10 but in subjects other than reading and math will be referred to as “Type B” teachers.) For teachers who don’t teach students who take statewide assessments or do not have baseline scores because their students are in kindergarten through grade 3, and grades 11 and 12 (“Type C” teachers), the Act mandates that “measurable learning targets must be established based upon the goals of the school improvement plan and approved by the school principal.” Id.

The Act also allows district superintendents to draw instructional personnel into the state formula by assigning them to an “instructional team” and to base those teachers’ evaluations on “the student learning growth of the instructional team’s students’ on statewide assessments.” Id. The Act does not define the term “instructional team.” Nor does it require that teams be structured in a way that actually allows teachers to work together on the instruction of the team’s students.

Following the passage of the Act, each of the Defendant school districts adopted an evaluation policy. See ECF No. 97-1. The Florida Department of Education (“DOE”) approved these policies, including those portions that allowed Type B and Type C teachers to be evaluated based on student scores on statewide assessments (namely, the FCAT). Id. at 7,12,19. The DOE aided in this process by providing the districts with state-approved VAMs that could be used to evaluate Type B teachers based on their students’ FCAT scores. ECF No. 86-3, at 96, 99, 101. With regard to Type C teachers, in each letter of approval, the DOE stated that it accepted the practice of evaluating Type C teachers based on “school-wide FCAT value-added results” on the grounds that such a practice constituted “learning targets based a link [sic] to student achievement goals in your school improvement plan as assessed by FCAT.” ECF No. 97-1, at 7, 12, 19. The DOE advised that “during the rule-making process for implementation of the ... Act, it is expected that the use of ‘learning targets’ and ‘instructional teams’ for evaluation purposes will be addressed and these changes are likely to result in the necessity for you to make adjustments to your evaluation plan for these teachers for the 2012-13 school year.” Id. It appears from the record, however, that all three Defendant school districts evaluated Type C teachers based. on school-wide FCAT results in the 2012-2013 school year, as well. See id. at 10, 17; ECF No. 88-1, at 78-80.

A. Alachua County’s Evaluation Policy

Alachua County’s policy provides that Type B teachers “will have their evaluation based on the scores of the students they teach using the reading associated VAM estimate.” ECF No. 86-3, at 71. For each Type B teacher, the state-approved VAM is used to calculate the effect that teacher has had on his or her students’ learning growth, as measured by their FCAT reading scores. Id. This measurement comes with an estimated error, which is used together with the calculated effect to compute nine data points: the calculated effect; the effect plus or minus one-half of the estimated error; the effect plus or minus the estimated error; the effect plus or minus one-and-a-half times the estimated error; and the effect plus or minus two times the estimated error.⁵ Id. The number of these data points that are above zero determines the teacher’s rating. Id. at 71-72. The policy provides that Type C teachers “will be evaluated based on the school component estimate of the VAM for reading.” Id. at 72. The policy employs the same nine-data point method for deriving a teacher evaluation rating from the VAM data. Id. at 72-73.

B. Escambia County’s Evaluation Policy

Escambia County’s policy provides that Type B teachers will be evaluated in the same manner as Type A or FCAT teachers. ECF No. 86-2, at 225 n. 3. That is, evaluations for Type B teachers are based on their students’ improvements in FCAT reading’ and math as measured by the state-approved VAM. ECF No. 86-1, at 114 ¶ 14. Each teacher’s rating is then determined by the percentage of their students who meet or exceed expected performance on the FCAT. ECF No. 86-1, at 112. The rating increases when more students exceed expected performance and decreases when students do not. Id. Type C teachers are rated in the same way, but since their students do not take the FCAT, their evaluations are based on school-wide average of FCAT improvement as calculated by the school component VAM. ECF No. 86-1 at 114-15 ¶¶ 12-13,15.

C. Hernando County’s Evaluation Policy

Hernando County’s policy provides that Type B teachers will be evaluated by their students’ improvement in FCAT reading and math as calculated by the state-approved VAM.⁶ ECF No. 86-1, at 124 ECF No. 86-2, at 222. Type C teachers are evaluated using school-wide FCAT improvement in reading and math as calculated by the school component VAM. ECF No. 86-1 at 123-24, ¶¶ V, W, Y; ECF No. 86-2, at 222. Because the VAM is a “co-variate adjustment model for estimating student learning growth,” (ECF No. 86-1, at 119 ¶ 7), it starts with an estimate of a student’s expected performance on the FCAT and identifies the effects that teachers or schools have had on the student’s learning by comparing the expected performance to the actual performance. ECF No. 86-1, at 11-13. Like the Escambia policy described above, a teacher’s rating is positively affected when students exceed expected performance and negatively affected when students fall short of expected performance. Id. at 13.

II. FACIAL CLAIMS AGAINST THE DISTRICT POLICIES

The gist of Plaintiffs’ facial challenge to the evaluation policies- is as follows: by mandating that teachers be evaluated on the basis of test scores earned by students who those teachers either (1) taught in subjects 'not associated with those tests or (2) didn’t teach at all, the evaluation policies are irrational and arbitrary and must be struck down under substantive due process and/or equal protection. ECF No. 86, at 13-21. The State Defendants — that is, the State Board of Education and Department of Education — argue that the districts’ policies are not irrational in allowing teachers to be evaluated partly on the basis of such student test scores because “professional teachers could positively impact all students at their schools” and “student learning in one subject [could] benefit ] students in other subjects.” ECF No. 84, at 20-24.

A. Substantive Due Process

Even though the policies at issue were enacted not by the legislature, but instead by the school districts and approved by the DOE, the policies are nevertheless “subject to constitutional infirmity to the same but no greater extent than if [they were enacted] by the state legislature.” Snowden v. Hughes, 321 U.S. 1, 11, 64 S.Ct. 397, 402, 88 L.Ed. 497 (1944). Thus the rational basis analysis to be applied to the policies is the same rational basis analysis already conducted for the Student Success Act. See ECF No. 111. The analysis requires this Court to identify “a legitimate government purpose — a goal — that the enacting government body could have been pursuing”; then ask “whether a rational basis exists for the enacting' government body to believe that the legislation would further the hypothesized purpose.” Restigouche, Inc. v. Town of Jupiter, 59 F.3d 1208, 1214 (11th Cir.1995) (quoting Haves v. City of Miami, 52 F.3d 918, 921-922 (11th Cir.1995)). If “a rational basis exists for the enacting govr ernment body to believe that the legislation would further the hypothesized purpose” it will pass rational basis review. Id.

For the evaluation of Type B teachers the inquiry is whether the DOE and school districts could rationally believe the district evaluation policies advance a legitimate state interest in increasing student learning growth. The answer is yes. First, it is rational to believe that a teacher can improve his or her students’ performance on statewide assessments even though the teacher is assigned to teach a different subject. Second, it is rational to consider the students’ improvement when evaluating the teacher’s performance. Finally, it is rational to believe that performance evaluations that take such improvement into account might incentivize better teaching, thus advancing the state’s interest in increasing student learning growth.

Similarly, for Type C teachers the evaluation policies pass rational basis review because the DOE and school districts could rationally believe that evaluating a teacher based on the performance of students who are not taught by the teacher but attend the teacher’s school advances a legitimate state interest in increasing student learning growth. First, as the State Defendants assert, it is rational to believe “that professional teachers could positively impact all students at their schools by fostering an encouraging learning environment, influencing and inspiring other teachers, exhibiting leadership, participating in school-wide strategic efforts, or otherwise making efforts to improve overall student learning.” ECF No. 84 at 21. Second, it would be rational to believe that school-wide improvements in student learning should be considered in a teacher’s performance evaluation. Finally, it is rational to believe that performance evaluations that take such improvement into account might incentivize better teaching, thus advancing the state’s interest in increasing student learning growth.

The fact that some of these beliefs might be “based on rational speculation unsupported by evidence or empirical data” does not doom the policies under the deferential rational basis standard. See FCC v. Beach Comm’ns, Inc., 508 U.S. at 307, 315, 113 S.Ct. 2096, 124 L.Ed.2d 211 (1993). So long as there is some rational basis that might conceivably support the policies, they will pass rational basis review regardless of whether the policies actually achieve their goals. The policies here pass rational basis review because the DOE and school districts could rationally believe that such evaluations further the state’s legitimate interest in increasing student learning growth.

B. Equal Protection

Plaintiffs allege in their complaint that the evaluation policies violate equal protection because they create “separate classes of teachers in Florida: those whose evaluations are based on student growth data for students assigned to the teacher in the subjects taught by the teacher, and those whose evaluations áre based on student growth data for students and/or subjects they do not teach.”⁷ ECF No. 48, Amended Complaint ¶ 124. Rational basis review applies to the classification because it does not involve suspect classes such as race, gender or national origin. Doe v. Moore, 410 F.3d 1337, 1345-46 (11th Cir.2005). Under rational basis review, a legislative classification violates equal protection only if there is no “ ‘reasonably conceivable state of facts that could provide a rational basis for’ it.” Id. (quoting FCC v. Beach Comm’ns, Inc., 508 U.S. 307, 313, 113 S.Ct. 2096, 2101, 124 L.Ed.2d 211 (1993)).

The classification Plaintiffs complain of has no hint of invidious discrimination but is rather a practical consequence of the lack of uniform measures available to assess student learning growth in all subjects and grade levels. Specifically, the FCAT VAM is available to provide an individual measurement of an FCAT teacher’s- contribution to his'or her students’ learning growth • in the subjects taiight in class. No comparable measurement is readily available for Type B and Type C teachers. As a temporary measure, the school boards have adopted the FCAT VAM to evaluate to Type B and Type C teachers even though the student growth data it provides is of very limited value for assessing these teachers.

A strong argument can be made that teacher evaluations should be postponed until tests are available for all courses and all subjects so that student growth can be assessed in a comparable manner for all teachers. The state, however,- is not prohibited under the equal protection clause from implementing teacher evaluation policies now to address what can be perceived as a pressing need to increase student learning growth. Equal protection does not force a legislative body to engage in reforms in a comprehensive manner, but instead allows reform to proceed “one step at a time.” Schilb v. Kuebel, 404 U.S. 357, 364-65, 92 S.Ct. 479, 484, 30 L.Ed.2d 502 (1972). Thus to be valid a legislative policy does not need to apply equally to all persons. Id. at 368, 92 S.Ct. 479. “ ‘The problems of government are practical ones and may justify, if they do not require, rough accommodations — illogical, it may be, and unscientific.’ ” Beach Comm’ns, 508 U.S. at 316 n. 7, 113 S.Ct. 2096 (quoting Metropolis Theatre Co. v. Chicago, 228 U.S. 61, 69-70, 33 S.Ct. 441, 443, 57 L.Ed.2d 730 (1913)). “ ‘[Ljogical appropriateness of the inclusion or exclusion of objects or person’ and ‘exact wisdom and nice adaptation of remedies are not required.’ ” Id. (quoting Heath & Milligan Mfg. Co. v. Worst, 207 U.S. 338, 354, 28 S.Ct. 114, 119, 52 L.Ed. 236 (1907)).

When the rational basis equal protection standard is applied here, the classification passes review because no invidious discrimination can be inferred from the classification; nor can it be said that the classification is without a “basis in practical experience” or that it is “wholly unrelated” to increasing student learning growth. Schilb, 404 U.S. at 368, 92 S.Ct. 479. The classification advances a legitimate governmental purpose by evaluating teachers based on the student learning growth measures that are available to the districts. Although the policies “impose special burdens upon defined classes” they may do so without violating equal protection because they are rationally related to “achiev[ing] permissible ends.” Id. at 368, 92 S.Ct. 479. It is not within the power of this Court to correct flaws in the policies so long as they are supported by a rational basis. “The Constitution presumes that, absent some reason to infer antipathy, even improvident decisions will eventually be rectified by the democratic process and that judicial intervention is generally unwarranted no matter how unwisely we may think a political branch has acted.” Beach Comm’ns, 508 U.S. at 314, 113 S.Ct. 2096 (quoting Vance v. Bradley, 440 U.S. 93, 97, 99 S.Ct. 939, 942-43, 59 L.Ed.2d 171 (1979)).

III. AS-APPLIED CLAIMS AGAINST THE DISTRICT POLICIES

Plaintiffs’ as-applied challenges to the district policies are based on the same alleged infirmity identified in their facial claims. The only difference between the claims is the breadth of the remedy Plaintiffs are seeking, which in their as-applied claims is limited to relief for the Plaintiffs themselves. See ECF No. 88, at 10 (“Plaintiffs explicitly pled both as-applied and facial claims, and Plaintiffs seek relief appropriate to those as-applied claims.”); see also Am. Fed. of State, Cnty. & Mun. Emps. Council 79 v. Scott, 717 F.3d 851, 863 (11th Cir.2013) (“[T]he scope of the relief requested ... determine[s] whether a challenge is facial or as-applied in nature.”)

There is nothing particular to Plaintiffs’ as-applied claims that would require an analysis different from the analysis of the facial claims addressed above. Indeed, the facial and as-applied claims are pleaded in the same counts and they involve the evaluation of Type B and Type C teachers using FCAT scores. As applied to the individual Plaintiffs in this case, the evaluation policies and the classifications pass constitutional muster because they are rationally related to a legitimate governmental interest.

IV. CLAIM AGAINST THE ALA-CHUA COUNTY POLICY

In their final claim, Plaintiffs contend that Alachua County’s teacher evaluation policy violates state law, specifically § 1012.34(3) of the Act, because Irby Elementary School teachers have been evaluated on the basis test scores of students at Alachua Elementary School.⁸ This Court declines to exercise supplemental jurisdiction over this claim because Plaintiffs’ federal claims have been resolved on summary judgment. The Eleventh Circuit has encouraged district courts to dismiss state law claims when all of the federal claims are resolved prior to trial. See Raney v. Allstate Ins. Co., 370 F.3d 1086, 1089 (11th Cir.2004).

V.CONCLUSION

The unfairness of the evaluation system as implemented is not lost on this Court. We have a teacher evaluation system in Florida that is supposed to measure the individual effectiveness of each teacher. But as the Plaintiffs have shown, the standards for evaluation differ significantly. FCAT teachers are being evaluated using an FCAT YAM that provides an individual measurement of a teacher’s contribution to student improvement in the subjects they teach. The FCAT VAM has been applied to Type B teachers as well, but perversely it can only measure student improvement in subjects not taught by the Type B teacher. For Type C teachers the FCAT VAM has been applied as a school-wide composite score that is the same for every teacher in the school. It does not contain any measure of student learning growth of the Type C teacher’s own students. To make matters worse, the legislature has mandated that teacher ratings be used to make important employment decisions such as pay, promotion, assignment, and retention.⁹ Ratings affect a teacher’s professional reputation as well because they are made public — they have even been printed in the newspaper. Needless to say, this Court would be hard-pressed to find anyone who would find this evaluation system fair to non-FCAT teachers, let alone be willing to submit to a similar evaluation system.

This case, however, is not about the fairness of the evaluation system. The standard of review is not whether the evaluation policies are good or bad, wise or unwise; but whether the evaluation policies are rational within the meaning of the law. The legal standard for invalidating legislative acts on substantive due process and equal protection grounds looks only to whether there is a conceivable rational basis to support them. For.reasons that have been explained, the State Defendants could rationally conclude that the evaluation policies further the state’s legitimate interest in increasing student learning growth. The same can be said of the District Defendants. Although they have not contested Plaintiffs’ motion for summary judgment and have not moved for summary judgment in their own right, they are entitled to judgment as a matter of law for the same reasons that summary judgment is granted for the State Defendants.

For these reasons,

IT IS ORDERED:

1. The State Defendants’ motion for summary judgment, ECF No. 84, is GRANTED.

2. Plaintiffs’ motion for summary judgment, ECF No. 86, is DENIED.

3. The Clerk shall enter judgment in favor of all Defendants against Plaintiffs on Counts One, Two, Six, and Seven, which have been dismissed, see ECF No. Ill; and on Counts Three, Four, Five, Eight, Nine, and Ten, on which summary judgment has been granted.

4. This Court declines to exercise supplemental jurisdiction over the remaining state law claim in Count Eleven.

1 . The school boards of Alachua County, Es-cambia County, and Hernando County did not respond to Plaintiffs' motion for summary judgment. This Court cannot, of course, enter summary judgment “on the mere fact that the motion was unopposed, but, rather must consider the merits of the motion.” United States v. One Piece of Real Property, 363 F.3d 1099, 1101 (11th Cir.2004). As an aside, this Court recognizes the impossible position in which the school boards have been placed and this Court is not unsympathetic to their plight. The school boards are attempting to implement an unfunded mandate. At a hearing on April 9, 2014, representatives of the school boards made plain that a conscious choice was made to abandon an aggressive defense so as to preserve the limited resources of the school districts. It is not lost on this Court that the school districts have been forced to pull experienced educators from the classroom to create tests to measure student growth, an action which ironically reduces student growth by removing experienced educators from the classroom.

2 . This Court entertained argument on the parties’ cross-motions for summary judgment on April 9, 2014. Noting the parties disagreed with respect to the application of the law to the facts, this Court specifically asked the parties whether there were any disputed issues of material fact. Neither the Plaintiffs nor the State Defendants identified any disputed issues of material fact which would preclude the entry of judgment.

3 . The reasoning behind the inclusion of part of the school component in each teacher's individual rating is discussed at length in American Institutes for Research, Florida Value-Added Model Technical Report 4-6 (2011). ECF No. 86-2, at 73-75. In short, it was decided that “some of the school component is a result of teacher actions within their schools and that [teachers] should receive some credit in their overall value-added effects.” Id. at 75.

4 . Section 1012.34(7)(e), Florida Statutes expires July 1, 2015.

5 . For the 2011-2012 school year, Alachua County used a slightly different nine-data point system. See ECF No. 86-3, at 56-57.

6 . Hernando County has been developing standardized tests to measure student learning growth in all courses not tested by the FCAT. ECF No. 86-1, at 123 ¶ S. For the 2013-2014 school years, tests were available for some courses. Id. They are expected to be available for all courses by the 2014-2015 school year. Id.

7 . Plaintiffs make a different equal protection argument in their motion for summary judgment. They argue that using the FCAT VAM to classify non-FCAT teachers as "highly effective,” "effective,” "needs improvement,” arid "unsatisfactoiy” under the teacher rating system violates their right to equal protection. ECF No. 86, p. 15. This equal protection classification is not pleaded in the First Amended Complaint, see ECF No. 48, so Plaintiffs cannot make this argument. Even if the classification had been pleaded, however, the argument is unavailing. "The equal protection guarantee has nothing to do with the determination of whether a specific individual is properly placed within a classification” and concerns "whether a classification is properly drawn. It is the guarantee of procedural due process that determines what process is necessary to find that an individual falls within or outside of a specific classification.” Ronald D. Rotunda and John E. No-wak, Treatise ori Constitutional Law — Substance & Procedure, § 18.2(a) (3d ed.1999).

8 . Plaintiffs note that for the 2013-2014 school year, Alachua County intends to evaluate Irby Elementary teachers using their students' performance on Discovery Education Early Skills Assessment Benchmark Tests. They contend that this claim is not moot because the new evaluation policy has not been finalized and because Irby Elementary teachers still need a remedy for the harm caused by the first two years of evaluations. See ECF No. 86, at 25 n. 19.

9 . The individual plaintiffs have a separate remedy to challenge an evaluation on procedural due process grounds if an evaluation is actually used to deprive the teacher of an employment right and if processes to challenge the deprivation at the state level are inadequate. See McKinney v. Pate, 20 F.3d 1550, 1560 (11th Cir.1994).