I've always been curious as to why everyone continues to use interviews as a primary means of assessing candidates. Managers don't like to do them, candidates literally hate them, and as a predictive indicator of performance, they stink!
# J! N, h' y2 E: s. k  h* A3 i) @7 o- `. }4 V  F
"Interviews are a terrible predictor of performance."
0 ]0 M8 R# M; N0 h2 s0 D7 F" z  R6 J# q8 T8 e, {5 A+ g
That quote isn't an opinion, it's a statement of fact based on the past experience of Laszlo Bock, Google's Vice President for People Operations as quoted by Saul Hansell in the New York Times on January 3, 2007.
: J6 ^! Q7 X5 |$ Y" O& i; ]+ v3 I; g0 x
The validity of interviews is regularly criticized by corporate professionals and thought leaders who have studied the success of interviews for predicting performance. Despite this criticism, their position as the foundation of most organizations' assessment system remains strong. 7 z- i# ~) j9 d. E" y

6 l0 |- N0 M3 G" qHistorically, there is plenty of data to demonstrate that interviews possess a predictive capability akin to that of a Ouija board. Most academic studies these days are not commissioned to test validity, but rather prove validity. As almost any PhD candidate can attest, nearly any research study can be crafted to produce exactly the results vendors with a vested interest in the outcome need to bolster their sales collateral. Rather than diving deep into the problems of studies conducted to prove validity, I've decided to show you in some logical detail what makes interviews so weak.
$ ^& |/ A2 n: c0 |" v3 y) [% q* }! k% \; W
Stuck in the Dark Ages
) g( k, F9 X6 X, @Interviews are so common that literally millions are conducted each day with little concern for their accuracy or effectiveness. Bloodletting and drowning witches were also once common, but luckily, we learned enough as humans to move past those practices.
2 c& s$ m- T& j8 v6 k. k$ n
1 a0 g7 n& b' OUnfortunately, the standard interview has barely changed. As the global economy continues to emerge, and firms adopt new practices that reflect the realities of operating a modern global organization, how they assess talent will become even more critical. & V: o& s& k  |, g7 \
5 i- y4 E8 g) D$ k2 L. N6 W) g0 w
Despite what HR directors trying to cover their behinds say, there are a multitude of cultural influences that affect how candidates from different geographic or ethnic backgrounds interview. In addition, due to decreased cycle times relating to the design, development, and obsolescence of goods and services, organizations today need less homogeneity and more diversity in thinking styles to drive innovation.
) W2 \" |: T6 N$ _5 l/ P
0 z) \* n9 `! V+ A9 Q4 M( jIf organizations are to continue being successful, interview systems will need to be robustly audited to ensure that they are not consistently screening out innovative individuals who tend to act "differently" during interviews. Unfortunately, only one in 1,000 firms even attempt to gather data on the accuracy or reliability of their interviews.

If you want to stop assuming that yours work, review this detailed list of the top-10 critical issues with interviews:
5 y$ O* w& Z5 v4 V; `1 e, k. P. T" z2 @
Some things should not be measured in an interview. The lack of agreement in advance as to what should and shouldn't be measured via an interview is a serious problem. It turns out that many things just can't be measured and shouldn't be measured during an interview. Unfortunately, that fact doesn't stop "independent-minded" managers from trying to measure inappropriate things, including body language, team skills, and intelligence.
1 Z6 E$ e& a* C: f- k  I$ c
  |* W) H: r5 x! mInterview questions are not directly related to needed skills. Each individual question needs to be tied to a specific job skill or knowledge. Unfortunately, most interview questions are developed independently and are not tied to any specific "required" skill or knowledge. , \2 n4 \; [+ |" O2 B) z
- _5 D+ D0 ?  r9 M# a1 M2 l
No script or plan. There is no interview "script" prepared for most interviews. As a result, no two interviews follow the same plan, making robust comparison of interview data difficult.
" Y3 x& ^- }$ s7 ]
. ?+ f- q* ^9 fNo weights. Interview questions are not weighted, so it is possible that the most important questions do not have the most influence on the final interview score outcome. 1 S. x! T* w( \3 o' |& w4 c
. Y5 l9 _4 B4 c: f0 R9 r1 [- \
No scoring methodology. Without a process to rate candidates' questions and factors consistently, it is difficult to ensure that decisions are based solely on the factors to be considered. # \" q1 r& E) f1 e
) W2 B) E4 n4 a$ b4 J" {$ X
No agreement on good answers. Without a pre-determined idea of what makes a good answer versus a bad answer, candidates' similar answers could elicit different "scores" from different interviewers. ; {" @8 c* ]2 |

& v8 U' ^  H$ B! [) A5 ]No accuracy check. The validity or accuracy of interviews is not checked by comparing whether those who received high scores during the hiring process turn out to be top "on-the-job performers" and vice versa.
, R+ m( l9 |& u: [" e5 x+ K4 l# O$ b9 e& h  E* k4 s8 A6 K0 v
Managers are not trained. Managers only receive cursory training; therefore, they don't know the pitfalls that can lead to bad interviewing and hiring results. Because "mystery shoppers" are not used, HR has no direct way of knowing what might be happening during an individual manager's interviews.
' ~) \2 ~' {0 h8 A4 Y+ ~# ?9 @; h( U. s. W+ p
Interviews are inherently misleading. From the start, the basic foundation of the interview is based on the premise that during the interview, candidates are acting "normally," which is unlikely because most candidates are scared to death before, during, and after interviews. In addition, we assume that candidates are telling the truth or that we make great lie detectors. I find that premise humorous because we all know that candidates routinely stretch the truth or tell us what they think we want to hear. The interview situation is by definition unreal: what happens during the interview might not be representative of what one would actually do on the job. The entire interview process is ill-conceived and designed to find candidates' faults, as opposed to gauging a demonstrated ability to do the job. 4 F. {' |* d; f$ W- q6 x

% f* V: p- [4 \# y  c# J1 bSubjective factors influence decisions. Numerous subjective factors (i.e., body language, race, sex, age, accent, height, handshake, dress, and physical disabilities) directly impact interview decisions. It's a fact that individuals from different backgrounds and different demographic categories receive higher interview scores than others.

Common Interview Process Errors
1 @- X8 C2 G+ FThe actual design of the interview process can cause many problems, including: - [! x5 ~4 m# p4 @

7 X: k( H" U9 `/ {: w0 I' N, c- q( n# w( g4 j1 r- _/ p  L% [/ I/ k
Behavioral interviews have inherent weaknesses. Asking candidates to describe how they handled a certain situation in the past is problematic. First, the candidate might describe an actual event in which they were involved but did not play the role for which they are taking credit. Second, if their verbal descriptions or their delivery happens to be "clumsy," their accomplishments will likely be understated (even though they actually did what they described). Third, the past is not always an accurate predictor of the future, and it is probably even less so in our fast-changing world.
) J# |! a6 m/ d" j7 S7 z7 T) n
' I6 j- g  F; U9 h0 r9 AThe interviewer. An interviewer's training, sex, age, biases, and experience dramatically impact their assessment of any candidate. All too often, interviewers act like junior psychologists; as a result, they make snap but inaccurate judgments. 7 E  ?& M5 v" E- A- Y: W$ f4 ]- L

$ j3 W; `: I' w0 t1 ~- U. kPredictability of the questions. Because most corporate interview questions come from behavioral interviewing books or purchased methodologies, candidates can often predict what questions will be asked. Interviews become even more ineffective if candidates can guess and practice the questions in advance.
6 t; h3 {+ U6 `5 O, {9 `9 G" Q% n0 y$ b  `" u0 s
Illegal questions and notes. Because interviews are not recorded, it's not unusual for illegal questions to "pop out," or candidates to inadvertently volunteer illegal or unnecessary information. Finally, the unfettered handwritten notes taken by interviewers can be embarrassing should they see the light of day in a court proceeding. , X5 v3 G: e0 }1 ?. R

9 b% C+ Y5 d. B7 j. WNo written record. Because most interviews are conducted without being taped or even with a written record, there is little evidence (should legal or EEOC issues arise) as to what actually occurred or didn't occur during interviews. " l1 ~- U+ `# C6 `+ X2 a/ Z: A# u1 Q
The time of day. Because multiple candidates are involved at different times of the day, it makes accurately comparing candidates' interview results difficult. For example, I become more critical in my assessments as the day goes on. / R9 f# p; N" s/ @  |# F1 O4 |

4 X: V7 [; a3 X  v1 oConsistent location. If the location of the interview is not consistent for all candidates, it might influence the candidate's assessment (i.e., lunch interviews produce different results than conference-room interviews).
0 B" W: p5 Q" @9 \; D" a5 S+ ^" J$ A8 G6 o% g7 W
Interview length. Interviews are often very short, making realistic assessment difficult. Due to time and business pressures, managers often eagerly make snap "first impression" decisions that are almost always inaccurate.
: e( t9 |- i5 _. y1 O) w' y( j  D7 s; x$ s8 f7 S
Fit. Many managers use interviews to measure an individual's "fit" with the team and the corporate culture. Unfortunately, there is little evidence that untrained managers can accurately assess this in 30 minutes.
- t% ^- N. b( O) ^. j& E7 ~
  C# M' {  N( Y" S5 h! ~) ?: j0 dPractice makes perfect. Top performers who have been in a job for a long time might be rusty in their interviews skills, which could negatively impact their scores. Unemployed candidates who have recently gone through numerous interviews might actually benefit from their extensive "practice" and do better than the rusty interviewees.
. E9 l8 B) z3 |: N' |* q: E; L6 X9 ^7 A8 D
Knockout factors. Many managers seem to make up subjective knockout factors that prematurely and often unfairly screen out qualified candidates.
/ s0 x2 V/ y3 R* O4 f) y* Y
+ |( G1 C' E* ?% YWhich jobs? Interview scores tend to vary based on candidates' interpersonal and communication skills. Yet many jobs don't require even average interpersonal skills. This means that some jobs (i.e., receptionist, salesperson, and recruiter) lend themselves to being assessed through interviews, while interviews for other jobs (i.e., welders, artists, and ditch diggers) may be horrible predictors of candidates' on-the-job success because they work alone. & I6 t' M$ u2 B1 A4 |
  K% ^; m+ v8 C0 P
Lack of technology. Interviews haven't changed much since the Stone Age. Unfortunately, very few firms have successfully integrated video or online interviewing into their processes. 2 z; _% W6 T7 e9 ^$ E

8 Z" W8 B/ k0 k: B! _Panel interviews. Panel or group interviews are often intimidating, making an already nervous candidate more nervous because of the number of people in the room hurling question after question.
5 O; V4 H1 s- ~' uThe order of the interview. If you are the first among all candidates in the interview process, you're less likely to be hired than if you are the last candidate. Unfortunately, where you appear in the order of interviews impacts your odds of success.
7 e1 x6 l8 i, r' j$ E3 B
" q8 x) K7 U: x  {) xHiring only for today and for this job. Hiring managers can be shortsighted and selfish. They frequently interview and hire based on their own short-term needs. Though companies should hire individuals for current and future needs, the reality is that most interview questions are not designed to assess competencies for other positions within the company.
; r3 o7 X) q: TAs you have probably begun to realize, interviews can be attacked on many fronts despite all of the logical reasons that exist to use them. While this list could go on and on, it is important to move on and tackle how interviews can become a barrier to the productivity of the recruiting function. / b; K: W  g) ]4 e# m
1 h  D. D, K+ `/ W
Next week: The conclusion of this two-part series will look at how the interview process routinely fails as a customer-service process, and how all of these issues impact a world-class staffing function.