Despite the formidable standard of review, see Anderson v. City of Bessemer City, 470 U.S. 564, 573, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985) (stating that "a finding of intentional discrimination is a finding of fact," to which we apply the clear error review), where we have a "strong, unyielding belief that a mistake has been made," it is our duty to reverse or remand, Powell v. Alexander, 391 F.3d 1, 7 (1st Cir. 2004) (quoting Fed. Refinance Co. v. Klock, 352 F.3d 16, 27 (1st Cir. 2003) ). I am of the firm belief that this is that rare case in which the district court's finding that there was no retaliatory animus was clearly erroneous. In light of the supported facts established by the district court, the weight of the evidence merited a closer look at the motivation of crucial actors in this story-namely that of the WES administration. Therefore, I respectfully dissent.
The district court found the "heart of the case" to be the December 8, 2014 meeting, the site of the first identified adverse employment action. Richard v. Regional Sch. Unit 57, 296 F.Supp.3d 274, 278 (D. Me. 2017). The district court found that "Superintendent Davis ran the District," and that Principal Bertinet and Vice Principal *62Roberts, the school's administrators, were simply "doing the perceived bidding of Superintendent Davis." Id. at 304-05. Accordingly, the district court imputed Superintendent Davis's motivation to all parties that took adverse employment actions against Richard. But, contradicting its own findings, the district court also noted that the school administration took adverse employment actions without the Superintendent's direction. In particular, the district court "[found] that Principal Bertinet and Vice Principal Roberts were following Superintendent Davis'[s] lead in exerting intense pressure on Ms. Richard from the time of the December 8, 2014 meeting to her April 2015 leave of absence and beyond." Id. at 304. But later recognized that "[t]he actions by WES administration [went] beyond merely following the Superintendent's wishes." Id.
The majority reads the latter district court statement as limited to Principal Bertinet's and Vice Principal Roberts' actions "against the children who were victims of T.K. and L.P.'s aggressive conduct." Id. However, a careful reading of the district court's order leads us to conclude that Richard was also subject to such actions "beyond" the Superintendent's directions. First, the district court clearly stated that "unsupported positions" were taken against both Richard, and T.K. and L.P.'s victims. Id. ("[T]he troubling aspect of their joining in [Superintendent Davis's] campaign against Ms. Richard is that it led them to take unsupported positions against not only Ms. Richard but also against the children who were victims of T.K. and L.P.'s aggressive conduct." (emphasis added) ). Given the context of this statement, the term "unsupported positions" refers to the mental state that set the foundation for Principal Bertinet's and Vice Principal Roberts' actions "beyond ... the Superintendent's wishes," as the district court described them immediately following examples of their actions taken against T.K. and L.P.'s victims. Id. 3 Second, the district court explicitly included Richard, along with one of T.K. and L.P.'s victims-B.D., as one of the actors in the uncertain events it recognized unfolded in RSU 57. Id. at 304-05 ("In sum, the evidence suggests that something was going on within the administration of RSU 57 concerning T.K., L.P., B.D., and Charlene Richard, but the Court is not sure what."). Thus, if the uncertain activity in RSU 57 was related to the WES administration's actions "beyond" the Superintendent's direction against T.K. and L.P.'s victims, as the majority seems to concede, it follows that Richard was also subject to such unsupported, and in her case adverse, action. Therefore, the district court should have, but did not, consider the Principal's and Vice Principal's motivations separate and apart from the Superintendent's. This was clearly erroneous in light of the district court's own findings.
*63We commonly look to certain indicators such as temporal proximity, awareness of the protected conduct, comments by decision makers, and differential treatment, as circumstantial evidence to guide a causation analysis. See Mesnick v. Gen. Elec. Co., 950 F.2d 816, 824 (1st Cir. 1991) ("[A] plaintiff should not be required to produce 'smoking-gun' evidence before prevailing in a discrimination suit. There are many veins of circumstantial evidence that may be mined by a plaintiff to this end."); Garayalde-Rijos v. Municipality of Carolina, 747 F.3d 15, 25 (1st Cir. 2014) (" '[T]emporal proximity' is merely one factor relevant to causation ... 'reinforced by other evidence.' " (quoting Trainor v. HEI Hospitality, LLC, 699 F.3d 19, 28 (1st Cir. 2012) ) ); Alvarado v. Donahoe, 687 F.3d 453, 459 (1st Cir. 2012) (stating that knowledge is required for a retaliatory motive). An analysis of causation, however, is one where "[c]ontext matters." Planadeball v. Wyndham Vacation Resorts, Inc., 793 F.3d 169, 178 (1st Cir. 2015) (quoting Burlington Northern & Santa Fe Ry. v. White, 548 U.S. 53, 69, 126 S.Ct. 2405, 165 L.Ed.2d 345 (2006) ).
The concrete evidence in this case shows that, even if Superintendent Davis was not aware of Richard's advocacy at the time of the December 8, 2014 meeting, Principal Bertinet and Vice Principal Roberts both had knowledge of Richard's actions and took adverse employment actions against this exemplary kindergarten teacher. Superintendent Davis's hostile treatment at the December 8 meeting may have been the first, but was most certainly not the only, adverse employment action taken. Yet, the district court's causation analysis only considered the temporal proximity between Richard's filing of the SAT requests for T.K. and L.P. and the December 8 meeting.
The hostility by the school administrators continued throughout the spring of 2015 without proven command or knowledge by Superintendent Davis. In the district court's own words, "Principal Bertinet and Vice Principal Roberts began micromanaging Ms. Richard's classroom, criticizing her asserted failures, and building a case for administrative sanction by reprimands." Richard, 296 F.Supp.3d at 279. The hostility was most evident in Principal Bertinet's misleading expressions to Richard's detriment. For instance, when Richard went to Principal Bertinet's office to inform her that T.K. had hit a female classmate, Principal Bertinet "sought to deflect blame from T.K. to B.D." Id. at 304. When B.D.'s parents inquired of the administration about why B.D. was being pulled out of the classroom and whether B.D. was being targeted by the administration, Principal Bertinet told them that it was "simply a friendly check-in" and attempted to blame Richard for miscommunication, despite no evidence of such. See id. at 295-96. In another incident, when Principal Bertinet emailed B.D.'s parents to alert them that T.K. stepped on B.D.'s hand, Bertinet insisted it was an accident despite Richard's report to the contrary. Id. at 296. More overtly, at the tape-recorded meeting with B.D.'s parents on April 3, 2015, Principal Bertinet openly lied to B.D.'s parents that "[t]hat [was] the first time that Ms. Richard ha[d] said that she felt that [B.D.] was being targeted," despite the fact that Richard had reported this bullying to the administration the previous day. See id. at 297-98. Three days after this meeting, Principal Bertinet placed Richard on a Corrective Action Plan. Id. at 304. These incidents occurred five months after the December 8 meeting. The motivation of the WES administration is therefore of crucial importance to the ultimate causation analysis of retaliation in this case.
*64Despite these and other independent actions by the school's administrators, the district court inexplicably and summarily ended its analysis into the motivation of the WES administration by stating "the evidence suggests that something was going on within the administration of RSU 57 concerning T.K., L.P., B.D., and Charlene Richard, but the Court is not sure what. The Court turns to Superintendent Davis." Id. at 304-05. The district court did not consider the temporal proximity of these school administrator's adverse employment actions to Richard's advocacy, but instead only focused on the temporal proximity of the December 8 meeting. While the administration's actions may only be circumstantial evidence of a retaliatory animus, the temporal proximity between Richard's advocacy and those actions can and should have been used in the district court's causation analysis. Sánchez-Rodríguez v. AT&T Mobility P.R., Inc., 673 F.3d 1, 15 (1st Cir. 2012) ("Very close temporal proximity between protected activity and an adverse employment action can satisfy a plaintiff's burden of showing causal connection...."); see, e.g., DeCaire v. Mukasey, 530 F.3d 1, 21 (1st Cir. 2008) (using temporal proximity to find factual error in court's ultimate conclusion regarding causation).
Further, in its causation analysis, the district court erroneously considered only whether Superintendent Davis's pretextual explanation for his behavior provided evidence of a retaliatory motive, but failed to consider the pretextual explanation that the school administration put forth. See Zapata-Matos v. Reckitt & Colman, Inc., 277 F.3d 40, 45 (1st Cir. 2002) ("[D]isbelief of the reason may, along with the prima facie case, on appropriate facts, permit the trier of fact to conclude that the employer had discriminated."). Although the district court did mention RSU 57's proffered motivation at the very initial prima facie stage, it wholly failed to review the administration's justification for their actions when considering whether Richard met her ultimate burden of persuasion. Compare Richard, 296 F.Supp.3d at 300-01 (explaining that the administration articulated a nondiscriminatory reason for its actions), with id. at 301-02 (looking only to Superintendent Davis's motivations in its causation analysis). As illustrated above, Principal Bertinet and Vice Principal Roberts both participated in and influenced the adverse employment actions, and thus their motivation-and whether their proffered nondiscriminatory justification was pretextual-was also relevant and should have been analyzed by the district court. See, e.g., Webber v. Int'l Paper Co., 417 F.3d 229, 236 (1st Cir. 2005) (analyzing animus of each participant that influenced an adverse employment action); Cariglia v. Hertz Equip. Rental Corp., 363 F.3d 77, 85-88 (1st Cir. 2004) (stating that the district court should have reviewed the animus of an employee that may have manipulated the information provided to the ultimate decisionmakers). The district court's failure to do so was clearly erroneous. An analysis of RSU 57's proffered reasons for its actions may have supported Richard's argument that the adverse actions were motivated by her advocacy. See Kosereis v. Rhode Island, 331 F.3d 207, 214 (1st Cir. 2003) ("Evidence that the employer's stated reasons for its actions are pretextual can be sufficient to show improper motive.").
When the record is reviewed in its entirety, the retaliatory inference is apparent. The more incidents Richard reported to the school administration, the more the administration intensified the pressure that it exerted on Richard instead of providing the requested aid for the students for whom Richard was advocating. See *65DeCaire, 530 F.3d at 21 (lack of hostility prior to protected conduct supports inference of retaliatory animus). The district court's failure to consider the school administrator's motivations to determine whether Richard had proven her retaliation claims was clearly an error.
Kindergarten teachers are the first in line for the identification and proper treatment of children with disabilities within our school systems. The education system, parents, and society as a whole rely on those kindergarten teachers to advocate for children's needs and to create a safe environment in the classroom. Teachers, however, cannot be expected to shoulder this weight alone, but need the support of our school systems to help those students develop essential skills that they will employ for the rest of their lives. Richard was an exemplary teacher, and in some instances "the only positive force" in these children's lives. Richard, 296 F.Supp.3d at 296. RSU 57 not only unjustifiably disrupted her efforts to provide support for the children, but waged a campaign of adverse employment actions in retaliation. I am therefore left with a "definite and firm conviction that a mistake has been committed." United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948). I respectfully dissent.
To support its conclusion that the school administration took no action beyond Superintendent Davis's wishes against Richard, the majority points out the district court's omission of a specific example, in this particular part of its order, of a hostile act by Principal Bertinet and Vice Principal Roberts against Richard. See Maj. Op. 57-58 (citing Richard, 296 F.Supp.3d at 304 ). However, as is discussed in detail below, Principal Bertinet responded in a manner hostile towards Richard in the events surrounding two of the three examples provided by the district court and alluded to by the majority. Despite playing no role in suggesting that B.D. was the culprit, Principal Bertinet sought to blame Richard for B.D.'s trip to the principal's office following the incident involving T.K. and a female classmate in which T.K. had hit the girl, Richard, 296 F.Supp.3d at 295-96. Furthermore, during the tape recorded meeting, Principal Bertinet openly lied to B.D.'s parents by stating that Richard had failed to report to her that B.D. was being targeted, id. at 297-98, thereby implying to the parents that Richard was not properly supervising her students.