Filed 1/11/18 Massey v. City of Long Beach CA2/4
NOT TO BE PUBLISHED IN THE OFFICIAL REPORTS

California Rules of Court, rule 8.1115(a), prohibits courts and parties from citing or relying on opinions not certified for publication or ordered published, except as specified by rule 8.1115(b). This opinion has not been certified for publication or ordered published for purposes of rule 8.1115.

IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA

SECOND APPELLATE DISTRICT

DIVISION FOUR




DAMIEN MASSEY,

Plaintiff and Appellant,

v.

CITY OF LONG BEACH,

Defendant and Respondent.
B269205

(Los Angeles County
Super. Ct. No. BC518178)




APPEAL from a judgment and order of the Superior Court of Los Angeles County, Michael L. Stern, Judge. Reversed.
Law Office of Jeffrey A. Sklar, Jeffrey Alan Sklar and Joe Ollinger for Plaintiff and Appellant.
Office of the City Attorney, Victoria A. Silcox; Alderman & Hilgers, Allison R. Hilgers for Defendant and Respondent.
_________________________________
A jury rendered a verdict in favor of appellant Damien Massey on his claims of race discrimination and harassment under the Fair Employment and Housing Act (Gov. Code, § 12940 et seq., “FEHA”). The trial court then granted respondent City of Long Beach’s motions for a judgment notwithstanding the verdict (JNOV) and for a new trial. We reverse and reinstate the jury’s verdict.

FACTUAL AND PROCEDURAL SUMMARY
Appellant, who is African American, was hired as a public safety dispatch trainee at respondent’s communications center in March 2011. The center handles 911 phone calls from the public and radio calls from police officers. Lieutenant Kenneth Rosenthal, who is Caucasian, had been in charge of the center since October 2010. At the time, Leslie Griggs, who is African American, was assistant supervisor and lead trainer.
Rosenthal did not make the decision to hire appellant, but nevertheless called appellant on the phone to offer him the job. According to Griggs, after the phone call Rosenthal told her, ‘“Damien is not going to make it.”’ Griggs thought Rosenthal’s comment was race-based—she assumed that although Rosenthal had not seen appellant, he had guessed his race from the way appellant spoke on the phone. Rosenthal did not recall making the comment.
Appellant graduated from a training academy led by Griggs and another assistant supervisor, Caroline Schlott. He was then trained by experienced dispatchers over several months. Public safety dispatchers employed by the center typically work 10-hour shifts, splitting their time between phone and radio calls. The call traffic varies in intensity depending on the shift and day, which is why trainees are assigned to different trainers at various points during the training. On a slow day, dispatchers may field five to 10 “priority 1,” or emergency “light and siren,” calls in a four-hour radio shift; on a busy day, the number of such calls may be 30 to 40. Trainees were expected to pass five “benchmarks,” at seven and nine weeks, and at three, six, and nine months. They were expected to be able to work independently, or “solo out” on the phones after nine weeks and on the radio after nine months. Upon passing the nine-month benchmark, trainees would be promoted from public safety dispatcher I to public safety dispatcher II. They remained on probation for a full year, and continued to be evaluated to assure they had not regressed. Estimates as to how many recruits made it through training varied from five percent to 50 percent.
The other four recruits in appellant’s class were Shaina Blalock, who was Caucasian; Christy LaMadrid, who was Hispanic; and Terri Atwell and Nona Taite, each of whom was African American. LaMadrid left the program shortly after she started. Blalock was promoted early, in October 2011, on Griggs’s recommendation. Taite and Atwell resigned in January 2012.
David Barrow was appellant’s first trainer, between May and July 2011. Barrow graded appellant’s performance on a weekly evaluation form, with scores ranging from 1 (needs improvement) through 2-3 (shows progress) to 4 (meets benchmark requirements). The scores were applied to skills divided into four categories (performance, knowledge, interpersonal skills, and job readiness). Barrow occasionally added plus signs to scores below 4 and supplemented the numeric scores with written comments. Appellant’s scores increased progressively from 2 to 4 on the phones, with consistent 4’s at the time appellant passed his seven- and nine-week benchmarks. Barrow’s written comments generally were positive and encouraging; he commended appellant’s continued progress and ability to follow instructions. Barrow noted appellant’s positive demeanor and punctuality, but also noted his lack of confidence.
Appellant’s nine-week benchmark deadline was extended by a week. At trial, Barrow testified he was not confident appellant could work independently on the phones by the nine-week benchmark and was skeptical about his motivation because appellant often would return last from breaks or arrive last for meetings; however, because appellant was not late, Barrow did
not include his concerns in the evaluations. After talking to Griggs, who was certain appellant was ready to solo out on the phones, Barrow felt comfortable to go along with her recommendation.
Appellant’s scores on the radio varied from 1 to 4. In July 2011, he was recorded as having passed the three-month benchmark ahead of schedule, but he was later recorded as having passed the same benchmark in September 2011. In the meantime, appellant was trained on the radio by Stephanie Graham, who is African American, as well as other trainers, including Jason Hunter. During this period, appellant’s scores were largely in the 2-3 range, with occasional 4’s. Plusses were rarely used. Graham noted that appellant had difficulty talking and typing at the same time and giving out information efficiently; she also noted his pauses on air, his occasional defensiveness, and his lack of punctuality. The comments of other trainers were more positive.
In September 2011, appellant was officially transferred to Hunter, who for the most part evaluated him on a weekly basis, with a couple of 2’s early on and an increasing use of 3+ and 4. In November 2011, appellant passed the six-month benchmark with perfect scores. Hunter and other trainers during this period observed appellant do well in slow and medium radio traffic. Hunter commented on appellant’s increased confidence and enthusiasm, and noted his punctuality and receptiveness to feedback. Since Hunter’s shifts did not usually experience the heaviest traffic, he did not think that appellant was ready to work independently on the radio.
Michelle Johnson, who is African American, trained appellant between November 2011 and early January 2012, when he was preparing for the final, nine-month benchmark, after which he would have been able to work independently on the radio. She and other trainers evaluated his progress largely through daily reports, and his scores during this period ranged between 3 and 4, with consistent 4’s by Johnson in December 2011, who noted only that appellant needed to improve his confidence and pick up his pace. During Johnson’s shifts, radio traffic was generally slow, with occasional moderate and heavy “spurts,” which appellant was able to handle. Schlott, who evaluated appellant on December 15, 2011, noted that he was confident and that officers in the field commended his voice control; she recommended that he be exposed to busier shifts on weekends to test his multi-tasking skills and performance under stress.
Johnson thought appellant was ready or “close to” ready to work independently on the radio, and sometime around December 2011, she recommended to Griggs that appellant be promoted out of the training program. Griggs agreed and made the recommendation to Rosenthal. Rosenthal refused to promote appellant, without giving a reason even though he had followed Griggs’s earlier recommendation to promote Blalock. Based on Rosenthal’s comment in March 2011 that appellant would not “make it,” Griggs believed Rosenthal’s refusal to promote appellant was based on appellant’s race. At trial, Rosenthal did not recall the conversations he had had with Griggs about Blalock and appellant. Nor could he recall why he had assessed appellant’s performance as not up “to standards” after reviewing his training records sometime in December 2011.
In November 2011, Rosenthal sent an e-mail to Griggs and Schlott, which he blind copied to Beverly Nieves, who worked in human resources and whom he was dating at the time. He wrote that he had reviewed Taite’s training book, had noticed “inconsistencies with our documentation,” and wanted to make “adjustments to our program.” He directed that all training books be reviewed.
At some point during the training, Rosenthal instructed trainers to evaluate trainees through daily observation reports and to check the underused boxes on the report forms: REM (remedial), NO (not observed), and NRT (not responding to training). At trial, most witnesses did not remember when these changes were introduced, but Schlott agreed with the estimate of appellant’s counsel that they occurred “four to six months” after the recruits graduated from the training academy, timing them between September and November 2011. Atwell testified they occurred at the end of 2011, “a couple of months” before she quit in January 2012. The NO box began to be regularly checked on appellant’s November and December 2011 daily reports by Johnson and substitute trainers. Neither of the other boxes was checked until 2012.
Griggs left her job for medical reasons in either December 2011 or on January 2, 2012. She later quit, claiming the work environment was hostile to African Americans. Before leaving, Griggs had complained to Sherriel Murry in human resources about Rosenthal’s treatment of her and “the black recruits,” including appellant. According to Murry, Griggs’s complaints were made after hours in a social setting and were not made with an expectation that a formal complaint be filed. Nevertheless, Murry memorialized them in January 2012 and forwarded the memo for further investigation. Only the first, heavily redacted, page of Murry’s memo was introduced into evidence. Nevertheless, Murry was allowed to answer that Griggs mainly complained that Rosenthal was undermining her supervisory authority in the training program, but some of her complaints were of race-based hostile work environment and discrimination by Rosenthal. Griggs’s complaints were not investigated. Murry testified that they should have been.
In mid-January 2012, after Griggs left, Barrow told Rosenthal that there were rumors about Griggs’s preferential treatment of the African American trainees. Rosenthal directed Barrow to write down the rumors. Barrow wrote a memo, in which he claimed trainer Michelle Wilk had complained that Griggs moved benchmark dates to accommodate the black trainees’ progress, and that another trainer had felt intimidated by Griggs into increasing Atwell’s scores. Barrow also claimed that Griggs had become close to appellant’s family, had kept his Sundays free as an accommodation, had matched him with the best trainers, had avoided placing him with white female trainers, and had advocated for him because there had not been a black male dispatcher before. According to Barrow, Taite was doing a substandard job, Atwell was mediocre, and appellant, who had “the motivation of a tack,” was just “sliding by.” Also included in the memo were claims that trainers were reluctant to bring their concerns to Griggs; that Blalock had been mistreated during the training academy; that it was suspicious appellant was still in training in January 2012 after a note had been circulated about soloing him out in early December 2011; and that Griggs had “bailed” on the recruits “when the crap hit the fan.” Barrow ended the memo with the following statement: “I hope this is what you wanted. I need to go take a shower now.”
At trial, Barrow testified the memo was based on conversations he had had with “junior personnel” at the time he wrote it, but could not provide additional information about the unattributed gossip and could not recall specifics. He explained he had felt uncomfortable writing the memo and had done so reluctantly, but he denied that Rosenthal had told him what to write. Rosenthal testified he was ordered by his commander to direct Barrow to record the rumors Barrow had relayed, and he sent Barrow’s memo to the commander. However, Murry testified Barrow’s memo should have been sent to Human Resources for investigation.
After Griggs left, Rosenthal took over the training program. He instructed trainers to attach printouts of calls to the reports for documentation purposes, and he discouraged the use of plusses on evaluations. According to Taite, trainers were no longer to give positive feedback; trainees were no longer to ask questions during calls, get copies of their evaluations, or write comments on them; and trainees were penalized even after correcting their mistakes.
Cameron Marien trained appellant from early January to early February 2012. At the time, the nine-month benchmark was set for February 4, 2012. Marien scored appellant’s performance on the radio mostly in the 3 range, with occasional 2’s and 4’s. Marien criticized appellant for his listening skills, looking at the trainer too much, and taking long pauses. He attached printouts of calls and often checked the NO and REM boxes. Rosenthal asked Marien to “reevaluate” the daily report for January 29, 2012, on which the scores and comments did not match. Marien revised the report to lower appellant’s scores from 3’s and 4’s to 2’s in several skill categories (accurate entry of details, verbal and listening skills, officer safety, and performance under stress), and added time estimates for remedial training.
Appellant complained to Marien that the changes on the report were unfair, but accepted them and promised to do better. Marien also revised the February 3, 2012 daily report. The revised report shows that appellant received 2’s for his ability to correlate information and prioritize radio traffic, and 3’s in multitasking and officer safety; he received remedial training in those categories. At trial, several witnesses testified supervisors sometimes asked trainers to “rethink” scores in order to align them with comments.
Schlott observed appellant on February 4, 2012, the nine-month benchmark date, and noted he was “getting close” to meeting that benchmark. But because he did not pass it by the set date, Rosenthal requested that his probation be extended so that he could get additional training. The Chief of Police and the Civil Service Commission approved the request.
Rosenthal assigned Kim Riley to train appellant in early February 2012. Riley prided herself on being a tough trainer—the last five recruits she had trained had not passed, and LaMadrid had quit. At trial, Riley testified that Rosenthal instructed trainers to use the REM (remedial training) and NRT (no response to training) boxes on the daily reports and to document thoroughly the trainees’ mistakes, but not their strengths. Riley also stated Rosenthal suggested that she not look at appellant’s prior evaluations and that she document issues related to officer safety, but she denied Rosenthal significantly changed how she trained or how she documented appellant’s performance.
Appellant had disagreements with his new trainer from the first day, when Riley counseled him on the correct order of radio transmissions. Because he had missed his benchmark and was running out of time, he was concerned about getting low scores. When appellant complained to Schlott about what he thought were inaccuracies in Riley’s evaluations, she encouraged him to write comments in response. In his comments, appellant acknowledged he had made mistakes, but took issue with Riley’s assessments of his lack of understanding and the extent of his mistakes. Riley then told appellant Rosenthal did not want any more feedback from him. Appellant felt communication with his trainer had been shut down, and Riley confirmed in later evaluations that he no longer talked to her as much. At trial, Murry agreed that prohibiting appellant from commenting on the evaluations was contrary to policy.
Riley criticized appellant for being unable to keep up with moderate radio traffic; talk, type, and listen at the same time; take accurate notes; correlate and give out information efficiently; and prioritize calls. Officer safety was a recurrent issue she brought up. Under Riley, appellant’s scores ranged between 1 and 3, with frequent use of the NO, REM, and NRT boxes. She often claimed he was not responding to training. She also recorded ongoing concerns and disputes about his punctuality.
At trial, appellant testified Riley raised her voice and intimidated him. She penalized him for asking her questions, even though she took questions from Blalock, whose station was nearby. Riley acknowledged that even seasoned dispatchers make mistakes, and appellant testified she encouraged Blalock when the latter complained about mistakes she had made; yet, Riley was intolerant of appellant’s mistakes. Riley wrote up appellant for taking his lunch break late when he was on an emergency call. He claimed she had not told him when to take his breaks; she claimed she had. There also were disputes regarding his use of sick time.
Appellant testified there were inaccuracies, inconsistencies, exaggerations, and fabrications in Riley’s evaluations. He claimed Riley lowered his scores even when he followed her suggestions—for instance, she scored him down both if he asked officers to repeat something he did not hear and if he wrote down what he believed was said on the assumption that he understood it correctly. He claimed she told him both that he was not loud enough and that he was too loud.
While Riley noted that officers appeared frustrated with appellant, appellant testified he had been told officers called in with positive feedback about him. On one occasion, supervisors reported that officers had called in to complain about appellant. Lieutenant Phil Cloughesy put the officers’ complaints in an e-mail, stating that appellant gave out information in a piecemeal fashion and that on February 9 or 10, 2012, he provided a cross-street on a call involving a female officer fighting with a suspect away from her car, when he could have found the exact location of the car using “AVL” (automatic vehicle location). Riley’s notes regarding this call indicate it occurred on February 8, 2012, and that there was a dispute between her and appellant about whether he initially aired the cross-street correctly. However, Riley noted that by the time the assisting officers asked appellant to use “AVL,” “he had already done so.” At trial, Cloughesy acknowledged he had been “work friends” with Rosenthal for 15 or 20 years.
In mid-March 2012, appellant was transferred to trainer Michelle Wilk because he had complained that Riley was “unfair” or “hard” on him. Wilk considered herself a “hard” trainer, too. She testified she started all her trainees with low scores, so they could have room to improve, and denied keeping the scores “artificially low”; yet, some of her scores were lower than justified by her written comments. She would sit back and let trainees work alone; she let them make mistakes before offering assistance and discussing problems. Wilk noted appellant was having problems keeping up with heavy radio traffic, pausing before airing information, not following through on calls, and failing to return a warrant, all of which she claimed implicated officer safety. She also claimed he lacked “situational awareness,” which meant he did not see “the big picture.” His scores were initially in the 1-2 range, and progressed to 2’s and 3’s during the course of her training. Wilk did not score appellant when radio traffic was slow, and she gave him low marks (1’s and 2’s) on a night of heavy traffic, noting that he was unable to handle it. Yet, appellant was not concerned about his low scores because Wilk told him to ignore the evaluations as he was “doing fine.” Schlott, who observed appellant in mid-April 2012, reported he was able to handle busy traffic with few minor issues and responded well to feedback, but she noticed that he tended to take longer pauses and make more spelling errors after being “reprimanded.” Wilk stopped training appellant at the end of April 2012 because she was “very pregnant” and found the additional documentation Rosenthal required to be stressful.
Beginning in late April 2012, appellant was monitored by supervisor Runnels, and assistant supervisors Schlott and Graham. Rosenthal ordered them not to share their evaluation memos with appellant. At trial, Rosenthal testified the memos were ordered under POST, and denied that appellant was not supposed to see them. Runnels testified the supervisors spoke to appellant about their observations, and appellant acknowledged receiving bi- or triweekly evaluations with scores and comments. He testified that the supervisors’ instructions sometimes conflicted, and he had to adjust to each supervisor’s expectations during different shifts.
By the end of May 2012, the supervisors recommended that appellant be soloed out on the radio and promoted to dispatcher II, with the caveat that he would remain on probation and any inability to handle a radio assignment would result in a review of his probationary status and possible termination. The unsigned recommendation was retracted on June 4, 2012 due to mistakes appellant made on June 1. On that day, appellant failed to issue a “code red,” i.e. clear the channel for officers to transmit, on two calls involving a suspect with a knife; one of the calls involved a hatchet. Runnels denied that the retraction was influenced by a conversation with Rosenthal, who in turn testified he did not know if he had seen the supervisors’ recommendation and said the retraction was based on Runnels’ own observations.
Schlott was the supervisor who actually monitored appellant on June 1, 2012, and in her memo she wrote that he did not type in the hatchet information until she prompted him. She also noted that the sergeant on the call complained that appellant did not appear to have heard that a hatchet was involved and did not issue a code red. Appellant’s excuse at trial was that he did not know what a hatchet, or a machete (as he remembered the word used in the call) was, and he was told not to ask questions. Runnels, who did not sit with appellant, but listened in on the calls, testified appellant knew enough because the call was coded as “a man with a knife” type of call. Appellant confirmed that he aired the call that way. He also testified that in his experience units would advise if they needed a code red.
After two more weeks of monitoring, on June 17, 2012, the supervisors recommended that appellant not be promoted and instead be terminated because he had not consistently shown that he could work independently on the radio. The recommendations of Runnells and Graham describe appellant’s deficits as lack of “situational awareness” and failure to see the “big picture.” All three recommendations note that respondent’s dispatch center is one of the busiest in California, and appellant’s inability to make split-second decisions would be a liability.
Rosenthal and the three supervisors each denied that he had told them what to write in their final recommendations, and Rosenthal claimed he had told the supervisors to follow POST. Appellant impeached Rosenthal with his sworn declaration submitted in support of respondent’s motion for summary judgment, which contained the following ambiguous language: “I gave them each their final recommendations regarding Mr. Massey’s progress.” Rosenthal explained that “gave” meant “received,” and that he used the supervisors’ recommendations as a basis for releasing appellant from probation. Based on Rosenthal’s recommendation, appellant’s release was approved by Rosenthal’s commander and the chief of police.
In 2013, appellant sued respondent under FEHA, claiming race discrimination, harassment, failure to investigate complaints, and retaliation. Respondent’s summary judgment motion was denied, and the case was tried to a jury. Respondent filed motions in limine to exclude evidence of alleged race discrimination against Taite, Atwell, and Griggs. The court ruled the evidence would not be allowed unless it pertained specifically to appellant. Respondent also moved to exclude Murry’s testimony of her opinion that Rosenthal was a “bigot” or “racist.” The court ruled Murry could testify as to facts she knew, but not as to her opinion. On respondent’s motion, the court also provisionally excluded Barrow’s memo regarding rumors that the African American trainees received preferential treatment from Griggs and Murry’s memo regarding Griggs’s complaints against Rosenthal.
Respondent moved for nonsuit, which the court granted as to the claims for failure to investigate and retaliation. The discrimination and harassment claims were submitted to the jury, which returned a special verdict in favor of appellant on both claims, awarding him $693,492 in damages, $520,119 of which was for emotional distress and the rest for economic loss. After entry of judgment, the court granted respondent’s motion for JNOV and new trial, finding insufficient evidence of discrimination and harassment. Judgment for respondent was entered in December 2015. This appeal followed.

DISCUSSION
I
Appellant argues the JNOV motion was improperly granted. That motion “may be granted only if it appears from the evidence, viewed in the light most favorable to the party securing the verdict, that there is no substantial evidence in support.” (Sweatman v. Department of Veterans Affairs (2001) 25 Cal.4th 62, 68, citing Hauter v. Zogarts (1975) 14 Cal.3d 104, 110.) For purposes of a JNOV, substantial evidence includes evidence improperly admitted during trial since the proper remedy for evidentiary errors is a motion for a new trial or appeal. (Donahue v. Ziv Television Programs, Inc. (1966) 245 Cal.App.2d 593, 609–610.)
On appeal from a JNOV, we view the record in the light most favorable to the jury verdict, resolving all conflicts and drawing all reasonable inferences in favor of the verdict. (Carter v. CB Richard Ellis, Inc. (2004) 122 Cal.App.4th 1313, 1320.) To be reasonable, inferences “‘must be “a product of logic and reason” and “must rest on the evidence” [citation]; inferences that are the result of mere speculation or conjecture cannot support a finding [citation].’ [Citation.]” (Casella v. SouthWest Dealer Services, Inc. (2007) 157 Cal.App.4th 1127, 1144.) But “‘when two or more inferences can reasonably be deduced from the facts, a reviewing court is without power to substitute its deductions for those of the [trier of fact]. If such substantial evidence be found, it is of no consequence that the [trier of fact] believing other evidence, or drawing other reasonable inferences, might have reached a contrary conclusion. [Citations.]’ [Citation.]” (Byrum v. Brand (1990) 219 Cal.App.3d 926, 946.)
A. Discrimination
FEHA prohibits race-based discrimination “in terms,
conditions, or privileges of employment.” (Gov. Code, § 12940, subd. (a).) The plaintiff in a race discrimination case has the burden of proving that his or her “race was a substantial factor in the adverse employment decision. [Citation.]” (Horsford v. Board of Trustees of California State University (2005) 132 Cal.App.4th 359, 375 (Horsford).) Discrimination may be inferred from substantial evidence that the employer’s stated reasons for the employment decision are “unworthy of belief” and pretextual because the employer acted with a discriminatory motive. (Swanson v. Morongo Unified School District (2014) 232 Cal.App.4th 954, 965; Johnson v. United Cerebral Palsy/Spastic Children’s Foundation (2009) 173 Cal.App.4th 740, 759 (Johnson).) “[A]n inference of dissembling may arise where the employer has given shifting, contradictory, implausible, uninformed, or factually baseless justifications for its actions. [Citations.]” (Guz v. Bechtel National Inc. (2000) 24 Cal.4th 317, 363.) An inference of discriminatory motive may arise from circumstantial evidence, including disparate treatment of similarly situated employees (Iwekaogwu v. City of L.A. (1999) 75 Cal.App.4th 803, 815 (Iwekaogwu)), “stray” (ambiguous or remote) remarks suggestive of racial animus (Reid v. Google, Inc. (2010) 50 Cal.4th 512, 540–545), and “me too” testimony of mistreatment by other employees in the same protected class. (Johnson, at p. 760.)
The burden-shifting framework originating in McDonnell Douglas Corp. v. Green (1973) 411 U.S. 792 is used to analyze the parties’ respective evidentiary offerings at the early stages of litigation, but not after the case has been submitted to the jury; at that stage, the substantial evidence standard of review applies. (Horsford, supra, 132 Cal.App.4th at p. 375.) In Horsford, as here, the defendant proceeded on appeal to draw inferences favorable to its position and ignore inferences favorable to the plaintiffs; the defendant also unwound “the various strands of [the] plaintiffs’ evidence, discounting each strand, in an attempt to show that ‘the jury could only speculate that race may have played a part in [the adverse employment] decisions.’” (Id. at p. 377.) As the Horsford court explained, we must examine the record as a whole, considering the plaintiff’s evidence collectively and drawing from it all reasonable inferences the jury could have drawn. (Ibid.) “Circumstantial evidence of motivation does not unravel into ‘speculation’ when the evidence permits inferences that are ‘the product of logic and reason.’ [Citations.]” (Ibid.) In other words, individual strands of the plaintiff’s case that in isolation do not constitute substantial evidence of pretext or discriminatory motive may be sufficient when viewed together. (Johnson, supra, 173 Cal.App.4th at p. 759.)
Appellant complains that, because of his race he was denied promotion, subjected to unfair training standards, and eventually released from probationary status—all adverse employment actions. (See Gov. Code, § 12940 [discharge from employment or from training program leading to employment]; Yanowitz v. L’Oreal USA, Inc. (2005) 36 Cal.4th 1028, 1055 [action likely to impair employee’s “job performance or prospects for advancement or promotion”]; Iwekaogwu, supra, 75 Cal.App.4th at p. 815 [denial of promotion].)
1. Denial of Promotion
Griggs testified that Rosenthal followed her
recommendation to promote Blalock early, but denied the recommendation to promote appellant, without giving any reason. That appellant was treated differently from a similarly situated white trainee may be evidence of pretext. (See Iwekaogwu, supra, 75 Cal.App.4th at p. 817.) Even if Blalock had prior dispatch experience, about which nothing is known, that does not necessarily render her differently situated from appellant since at the time each recommendation was made both she and appellant were deemed ready to solo out on the radio by the lead trainer. Respondent argues that appellant was not ready, but the evidence on that point is in dispute. Johnson, his trainer, recommended him for promotion because she deemed him ready; Griggs, the lead trainer who reviewed his performance, agreed.
In February 2012, Rosenthal represented to the Civil Service Commission that he had determined appellant’s performance was “not up to standards” after having reviewed his training records in December 2011. But it is unclear whether Rosenthal had undertaken this review before he refused to promote appellant; nor did he testify why he believed appellant’s performance was deficient at that time. Rather, at trial, Rosenthal claimed not to recall Griggs’s recommendations or his reasons for not promoting appellant. An “uninformed, or factually baseless” justification for an employment action may be evidence of pretext. (Guz v. Bechtel National Inc., supra, 24 Cal.4th at p. 363.)
2. Changes to the Training Program
Respondent argues that the changes Rosenthal made to the training program affected only documentation, but also concedes that the evidence on that point is in conflict. There was testimony that the trainers were discouraged from giving positive feedback and encouraged to focus on documenting the trainees’ mistakes, but not their strengths. The trainees were not allowed to ask trainers questions, get copies of their evaluations, or write comments on them. The trainees also were penalized even after correcting their mistakes. These changes, and Rosenthal’s suggestions to individual trainers, had a negative effect on appellant’s training experience. His scores were lowered (sometimes artificially), his mistakes were overemphasized, and trainers began describing their communications with him as remedial training or nonresponsiveness to training. Appellant, who until then, had been receptive to feedback, began complaining about trainers’ exaggerating his mistakes, lowering his scores unfairly, and not allowing him to ask questions. Some of the changes, such as prohibiting the trainees from commenting on the evaluations and discouraging the trainers from giving positive feedback, contravened established policies and practices. The overzealous focus on the trainees’ mistakes and the absolute prohibition against asking questions also was inconsistent with the general culture of the communications center, where dispatchers were allowed to ask each other questions and admittedly made mistakes. It is reasonable to conclude that the changes in the program impaired appellant’s “job performance” and “prospects for advancement or promotion.” (Yanowitz v. L’Oreal USA, Inc., supra, 36 Cal.4th at pp. 1054–1055.)
Respondent’s argument that the evidence as to when the changes were made is “confusing at best” is unavailing because conflicts in the evidence were for the jury to resolve. The record supports an inference that Rosenthal introduced the harsher changes after Blalock was promoted and Griggs had left. The record is silent whether the changes remained in effect in subsequent training classes; hence, it may reasonably be inferred that only the three African American trainees were subjected to the changes. The inference is supported by appellant’s testimony that Blalock, who was working solo, but before March 2012 should still have been on probation, was allowed to ask questions of appellant’s trainer and was comforted when she made mistakes. (See Iwekaogwu, supra, 75 Cal.App.4th at p. 817 [disparate treatment is evidence of discrimination].)
Respondent argues that the changes were made in order to comply with POST, but the jury was not required to believe that justification since the standards Rosenthal referenced were not proffered. Moreover, as appellant points out, Rosenthal first mentioned his reliance on POST at trial and respondent thereafter relied on that justification in closing argument. However, a belatedly offered justification during litigation may be evidence of pretext. (See Reid v. Google, Inc., supra, 50 Cal.4th at p. 545, fn. 13 [reasons for termination first raised in motion for summary judgment were considered pretextual].)
Importantly, the Barrow memo and Murry’s testimony about Griggs’s complaints show there were racial tensions in the training program around the time Griggs left. The claims in Barrow’s memo indicate that he perceived the African American trainees as either incompetent or unmotivated, and that he suspected Griggs of having given them preferential treatment based on their race. On the other hand, Griggs complained that Rosenthal undermined her as a supervisor and subjected the African American trainees to disparate treatment. As Murry testified, Barrow’s and Griggs’s complaints should have been investigated, but they were not. An employer’s failure to investigate complaints of discrimination may be evidence of pretext. (Nazir v. United Airlines, Inc. (2009) 178 Cal.App.4th 243, 278–283.) Although the complaints were not made by appellant, they directly implicated him as one of the trainees who was subject to either favoritism or discrimination.
The jury reasonably could have inferred that race had become a factor in the training program. Because Barrow walked back his most serious claims at trial, and no other witness admitted spreading the rumors included in his memo, or confirmed his claims of preferential treatment, it is reasonable to infer that biased and unverified claims of preferential treatment in their early training prompted the harsh treatment of the African American trainees after Griggs left. (See Staub v. Proctor Hosp. (2011) 562 U.S. 411, 419–420 [biased employee may influence adverse employment decision unless nexus is broken by independent investigation].)
3. Release from Probation
Appellant’s release from probation followed another recommendation that he be promoted, which was quickly withdrawn, ostensibly because of appellant’s mistakes involving not issuing a code red on a single night The significance of these mistakes is in dispute. Had there been a serious issue involving officer safety, the supervisor could have taken over his radio channel, which is what was done if a trainee could not perform a task or endangered officers in the field. That did not happen. There also is evidence that the increased focus on officer safety in evaluations was suggested by Rosenthal, and that even trivial matters were evaluated in terms of officer safety.
The two additional reasons for appellant’s release from probation—his lack of situational awareness and inability to see the big picture—are suspect as they did not match any of the criteria listed on the evaluation forms and on the final benchmark checklist, and there is evidence that dispatchers do not generally acquire those skills until one or two years after working independently. Respondent argues the three supervisors independently reached their recommendations that he be terminated. However, in a sworn declaration Rosenthal had stated that he “gave” the supervisors their final recommendations. The jury was entitled to discredit Rosenthal’s explanation at trial that “gave” meant “received”, resolving the ambiguity in the declaration against respondent. Under the “cat’s-paw” theory of liability, the jury could then infer that the supervisors were a mere “instrumentality or conduit” through whom Rosenthal brought about the final adverse employment action against appellant. (See Reeves v. Safeway Stores, Inc. (2004) 121 Cal.App.4th 95, 114 [discriminatory actions may be channeled through innocent actors].)
Appellant’s training process was unusual because repeated recommendations to promote were not followed by promotion, but by adverse employment actions. While it is possible to view these inconsistencies as due to appellant’s uneven performance, they also support an inference that the reasons given for the adverse employment actions were disingenuous.
4. Additional Evidence of Racial Animus
Appellant’s overarching theory at trial was that the adverse actions taken against him were due to Rosenthal’s racial animus. The theory was based in part on two comments Rosenthal made to Griggs. In one, Rosenthal told Griggs that appellant was “not going to make it,” immediately after offering him the job over the phone; in the other, Rosenthal told her to “look in the mirror” when she questioned his actions during her own promotion to supervisor. Respondent attacks the first comment as giving rise to a speculative inference that it was based on appellant’s race, and the second as irrelevant because it had nothing to do with appellant.
In Reid v. Google, Inc., supra, 50 Cal.4th 512, the court rejected the proposition that “ambiguous remarks are stray, irrelevant, prejudicial, and inadmissible.” (Id. at pp. 540–541.) The court explained that the jury is tasked with ‘“disambiguating ambiguous utterances”’ and determining their weight. (Id. at p. 541.) While they “may not have strong probative value when viewed in isolation,” ambiguous remarks “gain significance in conjunction with other circumstantial evidence.” (Ibid.) They need not have been made in the context of the challenged employment decision in order to be probative. (Id. at p. 540.) Additionally, “me too” evidence of discriminatory behavior by the same decision maker against others in plaintiff’s protected group is “unquestionabl[y]” relevant. (Pantoja v. Anton (2011) 198 Cal.App.4th 87, 118.) Thus, although neither comment is race-based on its face, in conjunction with other evidence— that dispatchers made fun of the speech patterns of African American callers and that Rosenthal was friendly with white employees, but cold to black employees—the jury could rationally infer that Rosenthal made employment decisions based on race, and that he prejudged appellant based on his race, which he could have inferred from the way appellant spoke on the phone.
That Rosenthal requested an extension of appellant’s probation may raise an inference favorable to respondent on analogy with the “same actor” theory of nondiscrimination, but such an inference may be weakened by other evidence. (Cf. Nazir v. United Airlines, Inc., supra, 178 Cal.App.4th at p. 273 [evidence that discriminatory actor previously selected plaintiff for favorable treatment may raise inference of nondiscrimination, but should not be “divorced from its factual context” or given “undue importance”].)
In light of the conflicts in the evidence, the trial court should not have granted respondent’s JNOV motion.
B. Harassment
FEHA independently prohibits employers from harassing their employees “because of race.” (Gov. Code, § 12940, subd. (j)(1).) “[H]arassment focuses on situations in which the social environment of the workplace becomes intolerable because the harassment (whether verbal, physical, or visual) communicates an offensive message to the harassed employee.” (Roby v. McKesson Corp. (2009) 47 Cal.4th 686, 706 (Roby).) “When the harasser is a supervisor, the employer is strictly liable for the supervisor’s actions. [Citation.] When the harasser is a nonsupervisory employee, employer liability turns on a showing of negligence (that is, the employer knew or should have known of the harassment and failed to take appropriate corrective action). [Citation.]” (Id. at p. 707.)
Official employment actions may supply the evidentiary basis of the harassment claim if they are used as the means of conveying an offensive message or have such secondary effect. (Roby, supra, 47 Cal.4th at pp. 708, 709.) Furthermore, “acts of discrimination can provide evidentiary support for a harassment claim by establishing discriminatory animus on the part of the manager responsible for the discrimination. . . .” (Id. at p. 709.) Thus, “discrimination and harassment claims can overlap as an evidentiary matter. The critical inquiry when a court is deciding whether the evidence is sufficient to uphold a verdict finding both discrimination and harassment is whether the evidence indicates violations of both FEHA prohibitions, but nothing prevents a plaintiff from proving these two violations with the same (or overlapping) evidentiary presentations.” (Ibid.)
Although respondent acknowledges that, under Roby, discrimination and harassment claims may overlap, it insists that appellant has failed to prove his harassment claim because he was not aware of harassing conduct directed against other employees and did not complain of harassment or discrimination; no explicit or implicit references to his race were made; and there is no evidence that any of the negative performance reviews he received were race-related. We disagree. The evidence supporting an inference that Rosenthal’s changes to the training program, as well as instructions and suggestions to trainers and supervisors, were racially motivated also supports a claim of harassment. Respondent is liable for Rosenthal’s actions, and, in any event, Murry’s memo about Griggs’s complaints against Rosenthal, which implicated his mistreatment of African American trainees, should have placed respondent on notice that appellant’s complaints about trainers Marien and Riley in 2012 related to a racially hostile work environment.
For all these reasons, we conclude that the court erred in granting JNOV on the discrimination and harassment claims. Because the JNOV was improperly granted, it must be reversed.
II
Appellant separately challenges the order granting a new trial. “[Code Civ. Proc. s]ection 657 requires the trial court to ‘specify the ground or grounds upon which [a new trial] is granted and the court’s reason or reasons for granting the new trial upon each ground stated.’ The requirement of a written statement of reasons encourages careful deliberation by the trial court and creates an adequate record for appellate review. [Citations.] If the ground for a new trial concerns insufficiency of the evidence, the trial court must briefly recite the respect in which the evidence is inadequate, and identify the evidence that convinces the court that the jury should have reached a different verdict. [Citation.].” (Montoya v. Barragan (2013) 220 Cal.App.4th 1215, 1227.) While the trial court need not expressly weigh “‘each item of evidence supporting, or impeaching, the judgment,’” a statement of reasons that states ultimate, rather than evidentiary, facts is inadequate. (Scala v. Jerry Witt & Sons, Inc. (1970) 3 Cal.3d 359, 370.)
Here, the trial court’s order states a single ground for granting the JNOV and new trial motions—insufficient evidence: “Plaintiff has failed to submit substantial evidence to support a verdict that [he] was discriminated against based on his race through proof of the essential elements foundational to establish defendant’s liability. . . . Plaintiff has failed to show sufficient evidence of harassment based on harassing conduct because of his race or that there was a hostile work environment based on racially motivated harassing conduct.”
The order does not state the reasons for finding the evidence insufficient, nor does it state any evidentiary facts; it finds only that appellant has failed to prove the ultimate facts he was required to establish. Because the statement of reasons is inadequate, the new trial order cannot be affirmed on the ground that the evidence is insufficient. (See Oakland Raiders v. National Football League (2007) 41 Cal.4th 624, 637, citing Scala v. Jerry Witt & Sons, Inc., supra, 3 Cal.3d 359, 371.)
“If an order granting a new trial does not effectively state the ground or the reasons, the order has been reversed on appeal where there are no grounds stated in the motion other than insufficient evidence or excessive or inadequate damages. [Citations.] If, however, the motion states any other ground for a new trial, an order granting the motion will be affirmed if any such other ground legally requires a new trial. [Citations.]” (Sanchez-Corea v. Bank of America (1985) 38 Cal.3d 892, 905.) One of the grounds for a new trial in respondent’s motion was that the verdict was “against law” because it was unsupported by substantial evidence. Whether the jury verdict is against the law is reviewed in the light most favorable to the verdict—under the same standard of review that applies to JNOV motions. (Id. at pp. 906–907.) That ground does not support granting a new trial for the same reasons that granting the JNOV was improper.
A new trial may also be granted for “[i]rregularity in the proceedings of the court . . . or adverse party, or any order of the court or abuse of discretion by which either party was prevented from having a fair trial.” (Code Civ. Proc., § 657(1).) In its motion and on appeal, respondent argues a new trial should be granted for abuse of discretion based on the court’s erroneous evidentiary rulings, and for irregularity in the proceedings because appellant’s attorney repeatedly elicited barred testimony. (See Garcia v. ConMed Corp. (2012) 204 Cal.App.4th 144, 148 [attorney misconduct may be irregularity in proceedings and ground for new trial]; Townsend v. Gonzalez (1957) 150 Cal.App.2d 241, 249–250 [evidentiary rulings may be grounds for new trial if prejudicial to moving party’s right to fair trial].)
The evidentiary rulings respondent challenges pertain to testimony about discrimination and harassment by Griggs, Taite, and Atwell, and to memos by Barrow and Murry. This evidence was not categorically subject to wholesale exclusion. “Me too” evidence, showing that the employer discriminated against other members of a plaintiff’s protected class, is admissible to show discriminatory or biased motive or intent under Evidence Code section 1101, subdivision (b). (Pantoja v. Anton, supra, 198 Cal.App.4th at p. 112; Johnson, supra, 173 Cal.App.4th at p. 760.) It is probative in discrimination and harassment claims, especially when involving the same decisionmaker or harasser, and any prejudice may be mitigated by a limiting instruction to clarify the purpose for which the evidence was offered. (Pantoja, at p. 118.) Dissimilarities between facts related to other employees’ “me too” testimony and those asserted by plaintiff go to the weight, not admissibility, of the evidence. (Johnson, at p. 767.)
Additionally, evidence of complaints of discrimination and harassment and the employer’s failure to investigate them is relevant to pretext and knowledge, as well as to the separate claim of failure to prevent discrimination or harassment under Government Code section12940, subdivision (k). (See, e.g., Nazir v. United Airlines, Inc., supra, 178 Cal.App.4th 243, 278–283 [inadequate investigation is evidence of pretext]; Huston v. Procter & Gamble Paper Products Corp. (3d Cir. 2009) 568 F.3d 100, 107–108 [human resources employee’s knowledge of complaints of harassment is imputed to employer]; Dees v. Johnson Controls World Services, Inc. (11th Cir. 1999) 168 F.3d 417, 423 [failure to prevent harassment based on knowledge of prior complaints of harassment by others is relevant].) Contrary to respondent’s assumption, appellant did not offer Barrow’s memo for the truth of its allegations, as appellant’s theory at trial was that the allegations in the memo were in fact untrue. (See, e.g., Caro v. Smith (1997) 59 Cal.App.4th 725, 733 [out-of-court statement is admissible for non-hearsay purpose, such as to show notice, if hearer’s reaction, rather than truth of statement, is relevant].) Similarly, Murry’s memo of Griggs’s complaints regarding discrimination and harassment in the training program was relevant for non-hearsay purposes, such as notice, knowledge, and failure to investigate.
Respondent argues broadly that, during trial, appellant’s counsel committed misconduct by repeatedly soliciting previously excluded evidence, and the court at times reversed itself, depriving respondent of a fair trial. We disagree. The court in this case took a narrow view of the evidence, ruling that it was relevant only if it directly involved appellant; hence, its evidentiary rulings generally favored respondent. If anything, that was unfair to appellant, not the other way around. For example, due to the court’s exclusion of relevant “me too” evidence, only Griggs testified about her grievances at any length; Taite and Atwell did not, which allows respondent to argue on appeal that there was “nothing of particular import” in their limited testimony. The court did not allow appellant to impeach Murry with her prior testimony that she believed Rosenthal was a bigot, so there was no violation of the motion in limine prohibiting such testimony. Instead, Murry testified favorably to respondent that she received complaints about Rosenthal from both African Americans and non-African Americans, and therefore did not consider him to harbor a racial animus.
Respondent also complains about argumentative questioning by appellant’s counsel. The only cited example relates to a series of questions that suggested Rosenthal did not want a black male dispatcher because respondent’s police officers arrest mostly black men. While this series of questions was improper, respondent did not object until the third question, after which appellant’s attorney attempted to lay a foundation and abandoned this line of questioning when told that it exceeded the scope of direct examination. This limited example does not compare to the pervasive attorney misconduct during trial that was at issue in Martinez v. Department of Transportation (2015) 238 Cal.App.4th 559, 567, on which respondent relies.
We conclude that both the JNOV and new trial motions were improperly granted. Since we reverse and reinstate the jury verdict, we need not address appellant’s default argument that the court also erred in granting nonsuit on his two additional causes of action—for failure to prevent discrimination and for retaliation.

DISPOSITION
The judgment and order are reversed. Appellant is entitled to his costs on appeal.
NOT TO BE PUBLISHED IN THE OFFICIAL REPORTS




EPSTEIN, P. J.
We concur:



WILLHITE, J.



MANELLA, J.