1 CA-CV 25-0246 Nonprecedential Vacated Processed

ROBERTS v. STATE

Arizona Court of Appeals · Filed March 5, 2026

Authorities cited

Identified automatically; this list may not be exhaustive.

Opinion text

NOTICE: NOT FOR OFFICIAL PUBLICATION.
UNDER ARIZONA RULE OF THE SUPREME COURT 111(c), THIS DECISION IS NOT PRECEDENTIAL
AND MAY BE CITED ONLY AS AUTHORIZED BY RULE.

IN THE
ARIZONA COURT OF APPEALS
DIVISION ONE

CLINTON ROBERTS, et al., Plaintiffs/Appellees,

v.

STATE OF ARIZONA, Defendant/Appellant.

No. 1 CA-CV 25-0246
FILED 03-05-2026

Appeal from the Superior Court in Maricopa County
No. CV2019-005879
The Honorable John R. Hannah Jr., Judge Retired

CLASS CERTIFICATION ORDER VACATED; REMANDED

COUNSEL

Napier, Baillie, Wilson, Bacon & Tallone PC, Phoenix
By Michael Napier, Juliana B. Tallone
Counsel for Plaintiffs/Appellees

Arizona Attorney General’s Office, Phoenix
By Rachel M. Remes
Co-Counsel for Defendant/Appellant

Struck Love Acedo PLC, Chandler
By Daniel P. Struck, Anne M. Orcutt, Kristina R. Rood, Nicholas D. Acedo
Co-Counsel for Defendant/Appellant
ROBERTS, et al., v. STATE
Decision of the Court

MEMORANDUM DECISION

Judge Cynthia J. Bailey delivered the decision of the Court, in which
Presiding Judge Daniel J. Kiley and Judge D. Steven Williams joined.

B A I L E Y, Judge:

¶1 Defendant State of Arizona appeals the superior court’s order
certifying two subclasses under Arizona Rule of Civil Procedure (“Rule”)
23. For the following reasons, we vacate the order and remand for further
proceedings.

FACTS AND PROCEDURAL HISTORY

¶2 Plaintiff Clinton Roberts is a correctional officer employed by
the Arizona Department of Corrections, Rehabilitation and Reentry (the
“Department”). Plaintiff Donna Christopher-Hall is a former Department
correctional officer. Plaintiffs sued the State on behalf of themselves and
others similarly situated, alleging they are entitled to overtime
compensation pursuant to Arizona Revised Statutes (“A.R.S.”) section 23-
392 for time spent on uncompensated mandatory and extensive pre- and
post-shift activities.

¶3 According to Plaintiffs, when correctional officers arrive at
Arizona prison facilities, they are required to show identification, answer
questions about whether they are carrying contraband, empty all
possessions for a search, pass through a metal detector, and have their
possessions X-ray scanned. At certain facilities, officers must wait for a
tram to transport them to their assigned work unit. Officers may then be
subject to additional screening at their assigned work unit, including metal
detection and a search of their possessions. Before reporting to their
assigned units, officers must obtain keys, radios, and other gear from an
equipment depot. When they report to their unit, some officers are required
to receive a “pass-down briefing” from the officer leaving the post. At the
end of their shift, officers give a pass-down briefing to incoming officers,
return their keys and other equipment to the depot, and travel to the exit
gate. At some facilities, the officers are subject to a vehicle search before
they can exit. We refer to these activities collectively as the “pre- and post-
shift activities.” The Plaintiffs offered their own declarations that, at the

2
ROBERTS, et al., v. STATE
Decision of the Court

Lewis prison, these uncompensated activities took between one and three
hours per week.

¶4 The State offered evidence that not every correctional officer
is required to engage in all of the pre- and post-shift activities and disputed
Plaintiffs’ allegation that officers are not compensated for any time spent
participating in pre- and post-shift activities. It also submitted evidence
that Plaintiffs’ time estimates were overstated and that the time required
for the pre- and post-shift activities varies considerably by prison facility,
assigned unit, individual employee, day, and shift.

¶5 Plaintiffs sought class certification under Rules 23(b)(1)(B)
and 23(b)(3). The superior court concluded that common issues
predominated and granted Plaintiffs’ motion for class certification under
Rule 23(b)(3), but rejected certification under Rule 23(b)(1)(B). It found that
“all corrections officers in all locations undergo pre-shift screening,” that
“those who work at specific facilities are screened post-shift as well,” and
that “corrections officers are not paid for this time, as a matter of policy.” It
defined the class as “all former or current corrections officers” who are
required by the Department to engage in the pre- and post-shift activities
and divided the class into two subclasses:

(1) All corrections officers who are required to undergo security
screening before their shift either at the main prison gate, the
assigned work unit, or both.
(2) All corrections officers who are required to submit to a vehicle
search and pat down before leaving the prison complex at the
end of their shift.

¶6 The State timely appealed. We have jurisdiction pursuant to
A.R.S. § 12-1873(A).

DISCUSSION

¶7 The State argues the superior court abused its discretion by
certifying the class because Plaintiffs did not prove that common questions
predominate over individual questions as required by Rule 23(b)(3).1 We

1 The State also challenges the superior court’s Rule 23(b)(3) superiority

determination and the court’s formulation of the subclasses. Because we
vacate the class certification order on predominance grounds, we do not
reach these issues.

3
ROBERTS, et al., v. STATE
Decision of the Court

review the superior court’s class certification decision for an abuse of
discretion. Ferrara v. 21st Century North America Ins. Co., 245 Ariz. 377, 380,
¶ 6 (App. 2018). A court abuses its discretion when it commits an error of
law or if there is no substantial support in the record for its conclusion.
Flying Diamond Airpark, LLC v. Meienberg, 215 Ariz. 44, 50, ¶ 27 (App. 2007).

¶8 A plaintiff seeking class action certification must meet all the
requirements of Rule 23(a) and at least one of the requirements of Rule
23(b). Ariz. R. Civ. P. 23(a) & (b); Wal-Mart Stores, Inc. v. Dukes, 564 U.S.
338, 345 (2011)
.2 The plaintiff has the burden to prove by a preponderance
of the evidence that these prerequisites are satisfied, Ferrara, 245 Ariz. at
380, ¶ 6; Wal-Mart, 564 U.S. at 350-51; Olean Wholesale Grocery Coop., Inc. v.
Bumble Bee Foods, LLC, 31 F.4th 651, 665 (9th Cir. 2022), and the trial court
must conduct a “rigorous analysis” to decide whether the plaintiff has
carried that burden. Wal-Mart, 564 U.S. at 350-51.3 This may require the
court to “probe behind the pleadings,” such that its class certification
analysis may overlap with the merits of the underlying claims to the extent
necessary to determine whether the Rule 23 prerequisites are satisfied. Wal-
Mart, 564 U.S. at 350-51; Olean, 31 F.4th at 667 (internal citation omitted).

¶9 The superior court found that Plaintiffs had established all of
the Rule 23(a) requirements: numerosity, commonality, typicality, and
adequacy. It found that the case was not suitable for class action treatment
under Rule 23(b)(1)(B), but that Plaintiffs had satisfied Rule 23(b)(3) by
showing that “the questions of law and fact common to the entire class
predominate over any questions affecting only individual class members,
and that a class action is superior to other available methods for
adjudicating the controversy.” The State challenges this predominance
ruling, arguing the common question Plaintiffs identify (whether the pre-
and post-shift activities are “work” under A.R.S. § 23-392) is not capable of

2 Because Rule 23 is identical to Federal Rule of Civil Procedure 23, federal

cases construing the federal rule are authoritative. Ferrara, 245 Ariz. at 380
n.2, ¶ 6.

3 The State questions the superior court’s statement in its ruling that Rule

23 “should be construed liberally, and doubts concerning whether to certify
a class action should be resolved in favor of certification,” arguing that this
is no longer the correct standard under recent United States Supreme Court
decisions, including Wal-Mart. The superior court, however, recognized
the more rigorous standard discussed in Wal-Mart and purported to apply
it.

4
ROBERTS, et al., v. STATE
Decision of the Court

class-wide resolution, and the remaining inquiries necessary to resolve
Plaintiffs’ claims require individual proof.

¶10 The “predominance inquiry tests whether proposed classes
are sufficiently cohesive to warrant adjudication by representation.”
Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623 (1997). To decide this
question, a court must consider whether common questions—those where
“the same evidence will suffice for each member to make a prima facie
showing [or] the issue is susceptible to generalized, class-wide proof”—are
more prevalent or important than individual questions—those where
“members of a proposed class will need to present evidence that varies
from member to member.” Tyson Foods v. Bouaphakeo, 577 U.S. 442, 453
(2016)
(internal citation omitted).

¶11 The predominance inquiry “begins, of course, with the
elements of the underlying cause of action.” Erica P. John Fund, Inc. v.
Halliburton Co., 563 U.S. 804, 809 (2011)
; Neale v. Volvo Cars of North America,
LLC, 794 F.3d 353, 370 (3rd Cir. 2015). By understanding the elements of a
cause of action, the court can determine whether the evidence needed to
satisfy each element may be common to the class or must be individual to
each class member. Neale, 794 F.3d at 371. The mere existence of individual
questions does not defeat class certification, but Rule 23(b)(3) requires that
the common questions predominate over any individual questions. Id.

¶12 Here, to prevail on their claim for failure to pay overtime
compensation under A.R.S. § 23-392, each class member will need to show
not only that the pre- and post-shift activities constitute “work” under
Arizona law, but also that the amount of time they spent engaged in those
activities caused their total time worked to exceed 40 hours in a work week.
See A.R.S. § 23-392(A) (requiring law enforcement officers to be
compensated for time worked over 40 hours in a week).

¶13 The superior court ruled that no individual proof is needed
for a court to determine whether the pre- and post-shift activities constitute
“work.” It based this decision on the assumption that the relevant inquiry
is whether the pre- and post-shift activities are “primarily for the benefit of
the employer.” See Prendergast v. City of Tempe, 143 Ariz. 14, 19 (App. 1984)
(stating the test to determine whether time is compensable as work is
whether it was “predominantly for the employer’s benefit or for the
employee’s benefit”). Assuming the Prendergast test applies, we reject the
State’s argument that the superior court erred by ruling that this issue is

5
ROBERTS, et al., v. STATE
Decision of the Court

subject to common proof.4 It is not, as the State contends, necessary to
examine each class member about what they were doing during the pre-
and post-shift activities to determine whether those activities were
primarily for the benefit of the employer or the employee. Rather, the court
should look at the “totality of all the circumstances,” including the nature
of the work and the level of the restriction on the employee’s time to decide
if this time is compensable as work. Prendergast, 143 Ariz. at 20. This is not
an employee-by-employee decision, but one common to all class members
that does not require individual proof.

¶14 That is not the end of the inquiry, though, as no class member
would be entitled to relief unless they could show the amount of time they
spent on pre- and post-shift activities, that this time caused their total work
time to exceed 40 hours in a week, and the amount of overtime
compensation (if any) they are owed. See Tyson Foods, 577 U.S. at 464
(Roberts, J., concurring). The State argues these inquiries require individual
proof and predominate over the common question of whether the activities
are compensable.

¶15 The superior court ruled that no individualized proof would
be necessary on these questions because it found that all corrections officers
are required to undergo the pre- and post-shift activities and, as a matter of
policy, are not paid for this time.5 It wrote that a straightforward damages
calculation could be made based on the class members’ time records, which
would contain information about “where and when they went through
entry and exit screenings.” The court posited that the duration of the pre-
and post-shift activities in any given location could be proven by “sample
observations extrapolated over time” and an expert analysis of this
information would allow Plaintiffs to make a “sufficient collective case for
damages, even if the damages calculations also involve some
individualized inquiries.”

¶16 While this may be a reasonable approach, it is not one
proposed or supported by Plaintiffs, who had the burden on this issue.

4 See Roberts v. State, 253 Ariz. 259, 270, ¶ 45 (2022) (declining to decide

“whether the Prendergast definition of work remains operative, whether
subsequent developments may have altered it, or whether any Arizona
statutory provisions provide guidance on the meaning of work”).

5 There is evidence, however, that Department employees are compensated

for some of the time they spend on pre- and post-shift activities.

6
ROBERTS, et al., v. STATE
Decision of the Court

Haliburton Co. v. Erica P. John Fund, Inc., 573 U.S. 258, 275 (2014); see also Ariz.
R. Civ. P. 23(c)(1)(B)(iv) (requiring the superior court to describe the
evidence supporting the court’s determination); A.R.S. § 12-1871(B) (“If the
court finds that an action should be maintained as a class action, the court
shall . . . describe all evidence in support of its determination.”). In their
motion for class certification, Plaintiffs offered merely the conclusory
statement that “common questions of law and fact predominate over any
individual questions of law and fact.”6 Even after the State presented
evidence of the extensive individual issues that it believed would need to
be resolved to determine whether each class member was harmed and the
amount of their damages, Plaintiffs’ reply merely stated that they could
“obtain expert reports that deal with time and motion studies and
damages.” This in no way satisfied their burden to show the prerequisites
to class certification by a preponderance of the evidence. Haliburton, 573
U.S. at 275; Comcast, 569 U.S. at 33; Wal-Mart, 564 U.S. at 350-51; Olean, 31
F.4th at 665. Plaintiffs themselves appear to have recognized this
shortcoming, as before the superior court granted class certification,
Plaintiffs moved to supplement the record to include an expert declaration
of how to calculate damages on a class-wide basis.

¶17 It is also not apparent from the record that common questions
would predominate over individual questions, as the evidence shows that
not all correctional officers are subject to the same pre- and post-shift
activities.

• The Department operates nine prison facilities, many of which have
multiple units.7

• The volume of traffic and level of scrutiny at each facility varies by
complex, day, and shift.

• Security screening procedures also vary by prison facility and unit
and have changed over time. At seven prison facilities, correctional
officers proceed directly to their assigned units and pass through a
single security screening. Other facilities required an additional

6 The only common question Plaintiffs identified was whether the pre- and

post-shift activities “are considered time worked and therefore subject to
compensation pursuant to A.R.S. § 23-392.”

7 A tenth prison facility closed during the class period.

7
ROBERTS, et al., v. STATE
Decision of the Court

screening for some – but not all - units for part of the time period in
question.

• When correctional officers are screened, at some locations their
belongings are X-rayed while they walk through a metal detector.
At other facilities, officers’ belongings are manually searched while
they walk through a metal detector. If officers do not clear the metal
detector, they may be subject to additional searches. One unit uses
an electronic scanner instead of a metal detector.

• Trams are available at two prison facilities (previously three
facilities) to transport employees to some—not all—of the units at
those facilities. Correctional officers are not required to use the
trams and the amount of time it takes to ride the tram will vary by
prison facility, unit, day, shift, and the number of operating trams.

• Correctional officers who work eight-hour shifts are required to
attend a pass-down briefing when they begin their shift, but are
given fifteen minutes of overtime compensation for attending the
briefing. Officers who work ten or twelve-hour shifts do not attend
pre-shift briefings.

• Officers working an eight-hour shift are not required to provide a
post-shift briefing to the incoming officer for the next shift. If
working at certain posts, they may be required to wait for an
incoming officer or other staff member to relieve them, in which case
they may submit an overtime request or, at their supervisor’s
discretion, receive flex time.

• Not all correctional officers are required to obtain keys and radios at
the start of their shift; this requirement varies by unit and post.
Correctional officers working an eight-hour shift may obtain this
equipment during or after their shift briefing, in which case this time
is compensated. Correctional officers working ten or twelve-hour
shifts obtain this equipment after the start of their shift, and this time
is compensated. The time for obtaining equipment varies
significantly by unit, shift, post, and assignment.

• Several prison facilities do not conduct vehicle exit searches, but
even at those that do, the procedure varies by unit and, in some
cases, has changed over time.

8
ROBERTS, et al., v. STATE
Decision of the Court

• A correctional officer’s unit assignment may vary by shift, changing
the pre- and post-shift activities they are required to follow.

¶18 Given these significant individual factors, determining the
amount of time each class member spent on pre- and post-shift activities is
likely to require extensive individual inquiry and is not susceptible to
generalized, class-wide proof. See Tyson Foods, 577 U.S. at 454-60
(discussing when representative evidence may be used to establish class-
wide liability). A similarly extensive inquiry was required in a recent case
over the compensability of time spent in employee bag checks at Kirkland’s
stores. Miles v. Kirkland’s Stores Inc., 89 F.4th 1217 (9th Cir. 2024). There,
because bag check frequency and practices varied by store, employee, and
manager, the court would have to “inquire into the individual practices of
each store, manager and employee—something that would not allow a
court to resolve the issue in ‘one stroke.’” Id. at 1225 (citation omitted). The
level of individualized inquiry required in Miles sufficed to render the class
inappropriate for certification and was arguably less extensive than what
would be required here. Id. at 1226.

¶19 Nor does the evidence support the superior court’s
conclusion that the calculation of overtime owed to each class member is
straightforward.

• Correctional officers work eight-, ten-, or twelve-hour shifts. The
Department does not use time clocks to track correctional officers’
hours. Instead, officers track their own hours, record them to the
nearest 15 minutes, and submit them.

• Before fiscal year 2019, correctional officers manually recorded their
hours on printed forms and submitted any later corrections on
paper. Officers’ time has been recorded electronically since fiscal
year 2019. Timecard corrections continue to be documented on
paper.

• Requests for overtime have always been submitted and maintained
on paper. Correctional officers may elect to receive overtime as
compensatory time or cash payment.

• Determining the amount of overtime, if any, each class member is
entitled to for pre- and post-shift activities would require manual
review of payroll, personnel, training, and security records, which
may be stored in several different locations and in either paper or
electronic format.

9
ROBERTS, et al., v. STATE
Decision of the Court

• Not all class members were employed for the entire class period and
promotions, demotions, or rehiring may have affected their pay and
overtime rates. The overtime rate for each class member at any given
time during the class period would vary depending on if they
received high-risk assignment pay, a geographic stipend,
performance awards, merit increases, bonuses, and other incentive
pay. This information would have to be extracted from personnel
files with manual calculations and adjustments for the historical
overtime rates applicable to each individual.

• Correctional officers who were on leave or suspended during any
work week in the class period would not have engaged in the pre-
and post-shift activities for that week and therefore would not be
eligible for overtime for those activities. Correctional officers
working a light-duty assignment or partial days during any work
week in the class period may have been subject to some, but not all,
of the pre- and post-shift activities for that week. Determining an
officer’s leave or other special assignment status would require a
manual review of any paper time records and the electronic
timekeeping system.

• Correctional officers assigned to offsite posts or participating in
offsite training or specialty activities do not engage in the pre- and
post-shift activities. To determine whether a class member fits into
one of these categories would require manual review of shift rosters
or training records and/or verifying the information with the
supervisor.

¶20 These undisputed facts contradict the superior court’s
findings that all corrections officers in all locations undergo pre- and post-
shift screening, and that officers are not paid for screening time.
Additionally, determining whether each of the 8,000-plus correctional
officers employed by the Department during the class period is entitled to
damages for the time they spent on the pre- and post-shift activities and
calculating those damages will require extensive individual inquiry.
Although individualized questions related to damages do not prevent
certification under the predominance requirement, Tyson Foods, 577 U.S. at
453, certification is not appropriate when “determining liability for all class
members would require complicated individualized inquiries.” Castillo v.
Bank of America, NA, 980 F.3d 723, 731-32 (9th Cir. 2020) (affirming trial
court’s denial of class certification because plaintiff did not establish a
common method of proving the fact of injury and any liability). The
individual questions here concern not just damages calculations, but

10
ROBERTS, et al., v. STATE
Decision of the Court

whether each class member is entitled to damages. Plaintiffs offered no
evidence that those questions can be answered on a class-wide basis,
distinguishing it from the cases Plaintiffs cite, for the proposition that
individualized damages do not defeat class certification. See Vaquero v.
Ashley Furniture Indus. Inc., 824 F.3d 1150, 1153, 1155 (9th Cir. 2016) (plaintiff
introduced representative liability evidence and proposed to establish
damages through use of a survey, sampling evidence, or a special master);
Leyva v. Medline Indus. Inc., 716 F.3d 510, 514 (9th Cir. 2013) (plaintiffs
offered evidence that the only individual question—damages—could be
calculated using defendant’s computerized payroll and time-keeping
database). See also Olean, 31 F.4th 651 (plaintiffs offered a statistical
regression model showing price-fixing caused class-wide antitrust impact);
McDaniel v. Wis. Dep’t of Corr., 416 Wis.2d 516, 523-25, 538-39, ¶¶ 6, 12, 46-
50, 21 N.W.3d 749, 753-54, 759-60 (Wis. Sup. Ct. 2025) (plaintiff presented
an expert who proposed a method to determine damage calculations for
correctional officer screening activities).

¶21 Plaintiffs had the burden to produce evidence that liability
and damages calculations could be resolved as common questions or would
not predominate over the common questions if decided individually. They
failed to do so. Haliburton, 573 U.S. at 275; McDaniel, 416 Wis.2d at 523-25,
538-39, ¶¶ 6, 12, 46-50, 21 N.W.3d at 753-54, 759-60. Without such evidence,
the superior court abused its discretion in finding that Plaintiffs had met
their burden to establish predominance under Rule 23(b)(3). Ferrara, 245
Ariz. at 380, ¶ 6; Haliburton, 573 U.S. at 275.

CONCLUSION

¶22 For the foregoing reasons, we vacate the superior court’s
order certifying the class and remand for further proceedings.

MATTHEW J. MARTIN • Clerk of the Court
FILED: JR

11