dissenting.
I disagree with the majority’s analysis under Manzer’s second prong. Cf. Manzer v. Diamond Shamrock Chemicals, Co., 29 F.3d 1078, 1084 (6th Cir.1994). Singleton offered sufficient circumstantial evidence to persuade a jury that the Defendant’s proffered reasons for firing him were pretextual. As the non-moving party, “his evidence is to be believed, and all justifiable inferences are to be drawn in [his] favor.” Martin v. Toledo Cardiology Consultants, Inc., 548 F.3d 405, 410 (6th Cir.2008).
The majority does not properly apply the applicable standard of review here. “Without weighing the evidence or judging the credibility of witnesses, and drawing all inferences in favor of the non-moving party, a district court usually undertakes to determine whether there is sufficient evidence to permit reasonable jurors to find for the non-moving party.” Aparicio v. Norfolk & Western Ry. Co., 84 F.3d 803, 807 (6th Cir.1996). A reasonable juror, viewing the facts as well as any inferences in the light most favorable to Singleton, could find by a preponderance of the evidence that he is entitled to relief. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 252, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). I would therefore reverse the district court’s grant of summary judgment and would remand for further proceedings because there are three triable issues relating to Singleton’s termination. See Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986).
Singleton satisfied his burden to demonstrate that “an illegal motivation was more likely than that offered by the defendant.” Manzer, 29 F.3d at 1084 (emphasis in original). First, Singleton presented data logs that outlined the commonplace carelessness with which Select managed its pharmaceuticals and punished alleged diverters. Neither party disputes that a Dilaudid discrepancy prompted Select’s extensive investigation into Singleton’s nursing practices. However, whether the investigation was routine or exceptional is a genuine issue of material fact. Select contends that even the possibility of diverted narcotics required not only that the alleged diverter undergo drug testing, but also a thorough evaluation of her medical and attendance records. Select solicited, but never received or reviewed, Singleton’s drug test. Further, in the approximately seventeen month period between August 2, 2005, and December 31, 2006, the Med Dispense machine in question experienced nearly 343 discrepancies. In one instance, “Nurse Therisa Hughes ‘resolved’ a discrepancy with [Dilaudid] by noting[,] T don’t know what I did.’” There is no evidence that Therisa’s records were photocopied, sent to the corporate office, analyzed in an ongoing investigatory report, and made to serve as the basis for termination, as Singleton’s were. Indeed, pursuant to Singleton’s request for discovery, Select identified only one other employee of whom a drug test was requested. The record does not indicate that Select ever thoroughly reviewed her files — or that it reviewed them at all. Taken together, this evidence and silence could convince a jury that Singleton was singled out.
The significance of the temporal proximity between Singleton’s complaint and firing is likewise best left to a jury. Our *406precedents concerning this issue stand for the principle that timing matters. See, e.g., Lindsay v. Yates 578 F.3d 407, 421 (6th Cir.2009) (while temporal proximity “cannot alone prove pretext ... it can be used [as] ‘indirect evidence’ to support an employee’s claim of pretext”) (citing Asmo v. Keane, 471 F.3d 588, 593); Mickey v. Zeidler Tool and Die Co., 516 F.3d 516, 524 (6th Cir.2008) (“none [of our precedents] squarely stands for the proposition that temporal proximity alone may never show a causal connection”); McNett v. Hardin Cmty. Fed. Credit Union, 118 Fed.Appx. 960, 965 (6th Cir.2004) (finding causation when “only 13 days” separated protected activity from adverse action, reasoning that an “employer’s knowledge of the protected activity coupled with an adverse action occurring close in time can create an inference of causation where the particular circumstances strengthen the inference of causation”); DiCarlo v. Potter, 358 F.3d 408, 421 (6th Cir.2004) (twenty-one day lapse between protected activity and adverse employer action fell within close temporal proximity); Moore v. KUKA Welding Sys., 171 F.3d 1073, 1080 (6th Cir.1999) (“[t]he causal connection ... may be established by demonstrating that the adverse action was taken shortly after plaintiff filed the complaint [with the EEOC] and by showing that he was treated differently from other employees”).
Plaintiff was suspended fourteen days following his February 27, 2006 letter; his employment was terminated sixteen days later. In that letter, Singleton alleged that African-American employees were routinely “called off’ more frequently than their white counterparts. The complaint compared the frequency of “calls off’ between a white, male nurse with that nurse’s black and female co-workers and suggested that black employees were being called off at a disproportionate rate. According to Singleton, Black employees were called off twelve times while the white employees were called off only three. Director of Clinical Services Kim McGowen investigated the matter and submitted an explanatory report to Select’s CEO, Rick Daugherty, on March 10, 2006. She stated that while “[Plaintiff] is an excellent ICU nurse,” the “call off’ schedule properly reflected “staff that request first call off due to family issues ... [as well as] acuity, census, call-in’s, overtime [and] continuity of care.” The incident nonetheless prompted a series of e-mails among Select’s higher-ups. For example, McGowen referred to Plaintiff as a “disgruntled employee” who might “complain to corporate.” Daugherty, Regional HR Manager Josceylon Buchs (“Buchs”), and Human Resources Coordinator Janet Whitlock corresponded about their belief that Plaintiff had sued a former employer. Within days, Singleton was fired. Indeed, a “Confidential Incident Report” filled out by nurse supervisor Chris LeMaster on March 16, 2006, referred to Singleton as already having been terminated, prior to the conclusion of the investigation. But Select maintains that the firing decision was not made until March 27. The close temporal proximity would permit a reasonable juror to find it more likely than not that racial animus, rather than the results of the investigation, motivated Select’s decision to fire. Also, the majority’s assumption that “the reference to Singleton being ‘terminated’ is instead a reference to his having been suspended” would best be tested before a jury.
Finally, summary judgment was inappropriate because the parties dispute the central issue of which documents gave rise to the firing. Singleton had received solid performance reviews for much of his employment. Indeed, in their written evaluations of his work between July 9 and December 9, 2005, Singleton’s super*407visors described him as an “exceptional asset,” a “very strong ICU nurse,” and a skilled and “excellent ... clinician].” His July 9 evaluation awarded him the highest possible rating for pain assessment. And in her response to Singleton’s February 27, 2006 complaint, his supervisor characterized him as “an excellent ICU nurse.” The tides turned only a week after Singleton submitted a racial complaint. On March 5, 2006, the day the Dilaudid discrepancy was discovered, Select initiated an investigation which purportedly revealed “errors ... so extreme,” in the words of investigation report author Molly Persby, “that they [we]re beyond education.” Singleton’s “documentation regarding assessment of pain or anxiety” was deemed “terrible;” his “correlation between signing out narcotics and documenting administration,” “poor.” Select therefore argues that the results of the investigation prompted the firing. Singleton counters that despite his diagnosed dyslexia, he received reviews that would warrant retaining him. The majority opinion appears to make improper credibility determinations, weighs the evidence, and makes factual assessments that are best left for juries. If the facts, and all inferences, are drawn in Singleton’s favor, a reasonable fact-finder could find for Singleton by a preponderance of the evidence.
For the foregoing reasons, I do not agree that Select was entitled to summary judgment. At least three genuine issues of material fact remain, all of which should be left to a jury.