What happens to an organisation when people leave?

Today, one of your colleagues is packing the contents of his desk into a cardboard box. A few weeks back you were at a leaving do for someone in another department. On the personal scale, these can be sad events. But what does turnover augur for the organisation? Different studies show different things, for example that sales suffers, benefits or is unaffected by turnover rates. Contrasting hypotheses exist, but a meta-analysis by Tae-Youn Park and Jason D. Shaw takes us from the theories into the data. Let's start with the competing ideas.

The first position is that turnover disrupts how well an organisation does. More experienced workers perform better, an idea stemming from human capital theories. And people in an organisation for a while get to know each other, reducing transaction costs between them, an insight from social capital theory. Replacing people undermines these benefits and is costly.

Another view is that in a company with low turnover - people don't leave often - human capital is indeed accumulating. This means when people leave the organisation loses a resource, as per idea one. However, when turnover stands at a high rate, the company isn't accumulating appreciable human/social capital: it simply isn't a big part of how the organisation succeeds. Hence losses are fairly painless, and to boot the company is likely to be efficient at replacing staff, given it's such a routine issue. So low levels of turnover hurt, but the impact tails off at higher levels.

The third perspective turns this on its head: at low levels, turnover is actually useful, as it revitalises the workforce by eliminating misfits who harm performance. It's only when turnover gets too high, meaning fewer misfitting people are exiting, that costs exceed this benefit.

To decide between these possibilities and explore other factors described below, Park and Shaw identified 255 studies on the relationship between organisational performance and turnover using standard searches of databases. Through use of exclusion criteria, they arrived at 110 sources containing 371 correlations. Before performing analysis, correlations were coded according to a number of factors, from industry of organisation to methodological approach. We'll see these in a moment.

Across the studies, a significant negative effect was found of -.15, suggesting that a 1SD increase in turnover produces a .15 decrease in performance. And when turnover was higher, the negative relationship became if anything stronger. This aligns most strongly with the simple human/social capital predictions: turnover hurts at all levels.  But the relationships varied widely across the included studies, and it's important to understand what's behind this.  Here's the overview.

Turnover affects certain performance measures more than others

• Customer Satisfaction and Quality showed large effects
• Weaker effects found for employee attitudes, productivity and financial performance
• Generally the effects were greater when measuring performance soon after the turnover, rather than moderately far or far into future

Organisational Type and structure matters

More effects were found in

• smaller companies
• executive level samples
• industries such as healthcare and hospitality, coded as 'human-capital-centric'
• those with so-called 'primary employment systems' that focus around delivery through committed employees (instead of a transactional, control-system)

These all paint a consistent picture that when organisations rely more on human capital, turnover hurts more. A large company which differentiates itself through use of equipment/other resources, eg mining, is less hurt by departures, especially so if it isn't investing in people practices (such as training, development, or engagement strategies) that secure commitment.

Turnover type matters

Reduction in force (downsizing) and voluntary turnovers both showed a significant negative relationship to performance. Involuntary turnover (getting fired) showed a relationship not different from zero. As some theories suggest all turnover should hurt, and others suggest that involuntary turnover should help organisations by losing weak performers, it's important to take this away. Possibly the benefits of losing misfits are cancelled by the costs of rehire.

The author conclude that 'organizations must recognize that when turnover rates rise, their workforce and financial performance are at risk. They should
 search for strategies to mitigate and eliminate turnover, recognizing
 that lower turnover is always better.'

Park, T., & Shaw, J. (2013). Turnover rates and organizational performance: A meta-analysis. Journal of Applied Psychology, 98 (2), 268-309 DOI: 10.1037/a0030723

Status shifts in groups as extraverts disappoint and neurotics overdeliver

New research suggests that the higher status bestowed on extraverts in new groups may drop as their contributions become better understood. In the meantime, neurotic people may see their lower status improve.

Corrine Bendersky and Neha Parikh Shah investigated this in two studies. The first examined how 44 student teams working on MBA assignments over 10 weeks attributed status and competence to individual members. One week after forming, each member was asked to rate the other 3 to 5 members' group status - e.g., 'To what extent does each individual influence the group’s decisions?' - and expected level of contribution to the group. Ten weeks after meeting, the members again rated each other on status and (now actual) contributions.

Personality measures taken at the start of the course showed that more neurotic individuals received lower status ratings at the first measurement stage, but made gains at the second stage. Extraverts, meanwhile, received marginally higher initial ratings but these decreased by time two. The effects were small, possibly because researchers controlled for a wide range of measures including other personality factors, gender, cognitive ability and individual assignment grades, which may soak up what might otherwise be observed. Further analysis confirmed effects were not due to regression to the mean, as variability in ratings was similar across the two time points. Instead, it appeared that the status changes were due to neurotics being seen to contribute more than had been expected, and extroverts less than expected.

In a computer-based experimental follow-up, 340 participants rated a hypothetical colleague before and after seeing their response to a request to assist the participant on a task. Beforehand, the colleague received higher status ratings when described using extraverted keywords, and was considered more likely to assist than when they were described as neurotic. However, when the colleague responded with a generous offer of help, neurotics were rewarded with greater increases in ratings than extraverts. And when the colleague was tight with their time, they were punished more heavily when portrayed as an extravert.

Extraverts find it easier to make a rapid, positive impact, being assertive, dominant and talkative. But for ongoing contributions to a team, their demeanour may introduce problems. They can be poorer listeners, and less able to cope with others being proactive, leading to group competition. This means that initially high expectations can lead to disappointment. Meanwhile, the low self-belief and and sense of powerlessness associated with neuroticism can make it easy to dismiss their group value. But neurotics are keen to avoid social disapproval and not be seen as incompetent, making them motivated to prepare for activities and put effort in, over-delivering on their original promise.

Bendersky and Shah argue that we should recognise that status is responsive to these factors, updating dynamically as a group gets to know each other or experiences different conditions. And in terms of practical implications, they offer a warning:  'Managers may rely too heavily on extraverted employees, which could be problematic if these individuals become less appreciated group members over time. In contrast, introverted and neurotic employees may be underutilized because managers inaccurately assume they will be less effective team members. With experience working together, however, both types of people may be important and valued contributors to their teams.'

Bendersky, C., & Shah, N. (2012). The Downfall of Extraverts and Rise of Neurotics: The Dynamic Process of Status Allocation in Task Groups Academy of Management Journal, 56 (2), 387-406 DOI: 10.5465/amj.2011.0316

Further reading:
Judge, T. A., & Bono, J. E. 2001. Relationship of core self-evaluations traits—Self-esteem, generalized self-efficacy, locus of control, and emotional stability — With job satisfaction and job performance: A meta-analysis. Journal of Applied Psychology, 86: 80 –92.

Does it take two to tango to get win-win negotiation outcomes?

Negotiation training has been shown to lead to positive outcomes for parties on both sides of the table, identifying 'win-win' solutions and helping the wheels of the world turn more amicably. But many studies focus on consequences when both negotiators are trained using the same methodology, when the reality is that a counterpart from another organisation may be trained differently or not at all. What happens then? A study by Alfred Zerres and colleagues finds out.

The study recruited 360 business administration students as either a 'buyer' or 'seller' and gave each some information on the context to study. Participants then conducted an initial negotiation over computer chat with a counterpart (a buyer to their seller), each trying to achieve their assigned objectives. Pair performance was measured by combining the success of buyer and seller: the negotiation was designed to be non-zero-sum as some factors were framed as more valuable to the buyer than seller, and vice versa, making some settlements more 'efficient' or mutually attractive than others. For this first negotiation, pair performance was fairly inefficient, on average reaching just 61% of the best outcome for both parties.

After this, some participants were kept busy with a filler task, whilst others received negotiation training. The training introduced the idea of 'logrolling', trading factors that mean more to the person receiving than the one giving, epitomised by two children fighting over an orange when one is most interested in the juice and the other the peel. Trainees then had opportunities to work through an example and then try and extend the concept to some other cases.

After this session, all pairs reconvened for a second negotiation on a separate issue. In some cases both parties had received training, and their pair performance was significantly better. What about when just one party had been trained? They did just as well – but only if it was the seller who had gotten training. Seeking a causal mechanism behind this, the researchers had recorded the computer-chat negotiation interactions, and looked at the amount of active information exchange in each pair, specifically sharing their own priorities or asking the counterpart about theirs. This kind of information sharing is crucial for logrolling, and analysis confirmed that this was behind the better performance by pairs with a trained seller. In a third negotiation one month later, groups with a trained seller both were better at spotting logrolling opportunities but also at recognising entirely compatible factors such as a buyer wanting quick delivery and a seller wanting to clear their warehouse rapidly.

The buyer role in a negotiation is about loss-aversion (not being suckered into a bad purchase)  which encourages a vigilant but passive stance - 'convince me!' - putting the onus on the seller to push and shape the conversation. If the seller is focused on log-rolling, then that conversation will be more transparent and lead to better outcomes. In some cases, a purchasing organisation might benefit more by identifying vendors  that are trained in (virtuous, win-win) negotiation techniques than to invest in training their own buyers. Alternatively, organisations who do train buyers might want to go beyond providing new concepts and investigate whether changing motivational mindset could lead to better outcomes in negotiations that matter.

Zerres, A., Hüffmeier, J., Freund, P., Backhaus, K., & Hertel, G. (2013). Does it take two to tango? Longitudinal effects of unilateral and bilateral integrative negotiation training. Journal of Applied Psychology, 98 (3), 478-491 DOI: 10.1037/a0032255
 
Further reading:
Nadler, J., Thompson, L., & Van Boven, L. (2003). Learning negotiation
skills: Four models of knowledge creation and transfer. Management
Science, 49, 529 –540. doi:10.1287/mnsc.49.4.529.14431

Experienced job interviewers are no better than novices at spotting lying candidates

This post was written by Christian Jarrett and originally found on the BPS Research Digest blog.
 
For the penultimate round of the TV show The Apprentice, the competing entrepreneurs must face a series of interviews with a crack team of hardened executives. The implicit, believable message is that these veterans have seen all the interview tricks in the book and will spot any blaggers a mile off. However, a new study provides the reality TV show with a reality check. A team led by Marc-André Reinhard report that experienced job interviewers are in fact no better than novice interviewers at spotting when a candidate is lying.

The researchers filmed 14 volunteers telling the truth about a job they'd really had in the past and then spinning a yarn about time in a job they'd never really had. The volunteers were offered a small monetary reward to boost their motivation. These clips were then played online to 46 highly experienced interviewers (they'd conducted between 21 and 1000 real-life job interviews), 92 interviewers with some experience (they'd interviewed at least once), and 214 students who'd never before acted as a job interviewer. The participants' task was to identify the clips in which the interviewee was speaking truthfully about their work experience, and the clips in which the interviewee was fabricating.

Overall the participants achieved an accuracy rate of 52 per cent - barely above chance performance, which is consistent with a huge literature showing how poor most of us are at spotting deception. But the headline finding is that the more experienced interviewers were no better than the novice interviewers at spotting lying job candidates - the first time that this topic has been researched. Greater work seniority, having more work experience and having more subordinates at work were also unrelated to the ability to spot lying job candidates.

There was a glimmer of hope that interview lie-detection skills could be taught. Participants who reported more correct beliefs about non-verbal cues to lying (e.g. liars don't in fact fidget more) were slightly more successful at recognising which job candidates were lying (each correct belief about a non-verbal cue added 1.2 per cent more accuracy on average). Experienced and novice interviewers in the current study didn't differ in their knowledge about lying cues, which helps explain why the veterans were no better at the task. The more experienced interviewers were however more skeptical overall, tending to rate more of the clips as featuring lying.

"Our results provide the first evidence that employment interviewers may not be better at detecting deception in job interviews than lay persons," the researchers said, "although it is a judgmental context that they are very experienced with."

Although the main gist of the results is consistent with related research in other contexts - for example, studies have found police detectives are no better at spotting lies, despite their interrogation experience - this study has some serious limitations, which undermine the applicability of the findings to the real world. Above all, the study did not involve real interviews, which meant the participants were unable to interact with the interviewees in a dynamic manner.

Reinhard, M., Scharmach, M., and Müller, P. (2013). It's not what you are, it's what you know: experience, beliefs, and the detection of deception in employment interviews Journal of Applied Social Psychology, 43 (3), 467-479 DOI: 10.1111/j.1559-1816.2013.01011.x

Who pays the biggest price for managing emotional displays in the workplace?

Understanding workplace demands on our emotions is one of our popular topics. Recent research combines two issues we've reported on previously: surface acting, the form of emotional labour that involves expressing emotions you don't genuinely feel, and affect spin, a measure of the variability of a person's emotional experiences. The paper suggests that overall, surface acting places greater demands on people high in affect spin.

Daniel Beal and colleagues ran their study with 64 restaurant servers from seven US restaurants. At regular stages in a shift, participants used PDA devices to record states and behaviours they had experienced since the last collection stage. This included how much surface acting they had performed, stress and fatigue measures, and ratings of various emotional states (eg happiness, guilt). The latter was used to compute affect spin by determining each individual's 'emotional centre' and then establishing how much they varied from this centre across the study. Participation was for an average of 10 shifts, with four collections per shift (shift start, pre-rush, post rush, shift end).

The ultimate study outcome measure was fatigue, and the data confirmed the researchers' prediction that surface acting would affect this in two ways. Directly - effortful strategies use up psychological resources - and indirectly through heightened stress, as a consequence of body physiology being forced away from natural expressions. The researchers suspected that affect spin would further influence this story and put this to the test using a multi-level model of how acting, stress and fatigue interact, both for individuals with low-affect spin - meaning their emotions are relatively consistent and non-dynamic - and for those with high-spin.

High spin participants saw surface acting increase their fatigue to a greater extent than their low spin co-workers. We know that high emotional variability makes it difficult to anticipate what emotions will emerge; this may make it harder to wrangle these sudden states into shape - especially if the emotion to be masked is extreme.

Similarly, whereas low spin individuals find surface acting slightly stressful, those with high spin seem to be more affected. Beale's team predicted this, as high spin individuals are generally more reactive to emotionally resonant situations, exactly the situations where surface acting tends to be needed.

But there is a silver lining for high spin: although they feel more stress, they can shrug it off more easily. It's plausible that their nature leads them to experience more daily drama, and they have learned to cope with it as a part of life. weakening somewhat the path from stress to fatigue. Still, overall the high spin individuals ended up more fatigued from surface acting than their counterparts.

As emotional labour is part of so many jobs nowadays, in the burgeoning service industry and beyond, it's important to understand what the consequences are for employee wellbeing. Stress and fatigue are predictors of burnout and job turnover, so understanding risk factors for different kinds of people gets us a step closer to supporting them and helping the workplace to contain natural smiles, as well as forced ones.
 

Beal, D., Trougakos, J., Weiss, H., & Dalal, R. (2013). Affect Spin and the Emotion Regulation Process at Work. Journal of Applied Psychology DOI: 10.1037/a0032559

Wish you were here!" - how a postcard can help attract the best talent

In 2004, in Silicon Valley, Google posted a huge billboard ad featuring a mathematical problem. The answer led to a web address with yet another puzzle to crack. People who successfully followed this intellectual treasure hunt ended up being invited in for a job interview.

This is an extreme example of a recruitment  principle spelled out in a new article by psychologists in Belgium. They say that distinctive recruitment procedures are the secret to attracting more and better job applicants, especially in fields like engineering where competition for the best talent is intense.

Working with a Belgian technology company, Saartje Cromheecke and her colleagues sent out a real job opportunity to 1,997 potential applicants, around half of them via email (as is the industry standard), and half via a hand-written postcard depicting a coffee mug and a blank daily agenda. The email and postcard message featured the same layout and included the same written information and content about the job vacancy.

Sixty-two of the contacted engineers applied for the job - 82% of them had received the postcard, just 18% had received the email. Stated differently, only 1% of the engineers who were emailed actually applied for the job compared with 5% of those who received a postcard. This latter figure represents a high response rate for the field. Moreover, the respondents to the postcard tended to be better educated, consistent with the researchers' prediction that a recruitment message sent via a "strange" medium will be more likely to grab the attention of better-qualified personnel who aren't actively looking for new opportunities.

The researchers said that social cognition research has shown how we adopt mental "scripts" for different aspects of our lives. "... recruiting in a strange way that differs from what competitors are doing is likely to be inconsistent with recruitment scripts," they said, "enhancing potential applicants' attention, attraction, and intention to apply."

It's important to note, Cromheecke's team aren't saying that postcards will always be the answer. Rather, "this field experiment puts forth 'media strangeness' as a more general evidence-based principle, which recruiters might take into account when selecting media for communicating job postings."

This post was written by Christian Jarrett and originally found on the BPS Research Digest blog. 

Cromheecke, S., Van Hoye, G., and Lievens, F. (2013). Changing things up in recruitment: Effects of a ‘strange’ recruitment medium on applicant pool quantity and quality. Journal of Occupational and Organizational Psychology DOI: 10.1111/joop.12018

Accountability provokes more team-focused behaviours in leaders who are outsiders

Sometimes leaders epitomise the group they seek to lead, such as a former trucker heading a transport trade union. In other cases leaders are less prototypical; while they may have the attributes for the role, they 'come from outside'. How might leaders from these two moulds respond when the workplace demands more accountability for their actions?  A team led by Steffen Giessner of Erasmus University set out to know more, investigating the team-oriented behaviours that leaders engage in when they know they will be scrutinised by followers.

At first blush, the prototypical leaders might be highly responsive under conditions of accountability. After all, it's harder to justify treating yourself as special and above a group when you resemble them so closely; better to act for 'your people' and cement your position as 'one of them'. But a first experiment with 152 students suggested otherwise.

Here each participant was led to believe they were leading a virtual team of three followers, and had been selected on the basis of either being very group prototypical or group non-prototypical, according to their answers to a questionnaire. They were then to complete decision-making tasks by assigning analysis work to their followers and making the final call on what answers to provide. Better answers would score points - some for each follower, and more for the leader - with group combined scores and individual scores both leading to possible financial reward.

What the researchers cared about was how the leaders would carve up the points-pie when they were given the authority to do so themselves. Just before this decision, half the participants were told they would need to justify their reasons to the team, and meet with them face to face before the end of the experiment. In this high accountability condition, the non-prototypical leaders dropped the proportion of points they kept for themselves to a level significantly lower than the baseline set by the experimental rules up to that point. Without the accountability, they held on to the baseline number of points, or even a little more.

 Meanwhile the prototypical leaders showed an intermediate level of generosity across both conditions. Their team-oriented behaviours didn't alter when accountability was put on the horizon.  Giessner's team believed that this reflects the relative security that prototypicality provides: by nature part of the in-group, there is less pressure to try and prove it when under scrutiny.

The investigators followed this with a field study  of 64 leaders and 209 followers. Leaders self-rated their prototypicality and how much accountability was present in the job, as well as another factor: team identification. Giessner's team suspected that in reality, accountability may not motivate non-prototypical leaders when they don't care about being part of the team, such as an interim manager aiming to get their job done before parachuting out. This hypothesis was borne out: followers' ratings of their leaders team-oriented behaviours (such as willingness to sacrifice own time for the benefit of the team) were high for non-prototypical leaders under accountability, but only if team identification was also high.

To make sense of this, I think of the postgraduate-trained specialist leading the salt-of-the-earth law enforcement team: at the end of the day, do they consider themselves police too? If so, they are likely to respond to increases in accountability with visible team-focused behaviour.

Of course, this research doesn't address other reasons why you might demand scrutiny of leader decisions, such as keeping them honest or providing transparency to a wider audience and thus helping information exchange. But as a tool for encouraging team behaviours, this evidence suggests that accountability may be most potent when aimed at outsiders who care about being included.

Giessner, S., van Knippenberg, D., van Ginkel, W., & Sleebos, E. (2013). Team-Oriented Leadership: The Interactive Effects of Leader Group Prototypicality, Accountability, and Team Identification. Journal of Applied Psychology DOI: 10.1037/a0032445

Further reading:

Rus, D., van Knippenberg, D., & Wisse, B. (2012). Leader power and
self-serving behavior: The moderating role of accountability. The Lead-
ership Quarterly, 23, 13–26. DOI:10.1016/j.leaqua.2011.11.002