By Simon Dyson (De Montford University) and Sue Dyson (Middlesex University)
Have you ever seen an early production line in action? Although these systems increased production, they did so at the expense of the quality of the experience of the workers, who were positioned at the bottom of the occupational ladder. The workers then had no say over the flow and pace of work. They had no voice in the organisation of work, nor any view of the overall wider picture into which their contribution fitted. Early studies in the sociology of work suggested that, in such circumstances, we should expect the following. Workers will do the minimum work possible to get by. They will not care about the quality of work they produce. They will not use initiative to solve any problems. Work will be undertaken resentfully, or else some will delight in finding creative ways to deliberately undermine the work processes.
A US sociologist, Julius Roth, considered these ideas in the light of his experiences as a hired data collector on big research projects. He suggested that research workers – whom he called “hired hands” – behaved in the same ways as disenchanted production line workers: they did not care about the quality of the data they produced, just so long as they did sufficient to stay out of trouble with their line-managers. When we were commissioned by the Department of Health to undertake a large health services research project we anticipated that the 300 or so midwives who would be collecting data would be put in this same “hired hands” position. From their point of view midwives had no say in the topic that was being researched, which concerned pregnant mothers being screened to see if they were genetic carriers of two inherited chronic illnesses, sickle cell anaemia or thalassaemia. They had no voice in the design of the study, no say in the analysis of data, no role in how the study was written up and no influence on what policy might be implemented as a consequence of the findings.
Because we realized the invidious position the midwives were being placed in, we tried to ensure they could see what was at stake for them (briefly: complex and emotionally difficult work in probing the ethnic and family origins of a mother to ascertain if she was at risk of carrying genes associated with sickle cell or thalassaemia, emotional work that was far more difficult and time-consuming than either policy-makers or their managers were prepared to allow); we also tried to ensure their time was fully paid for, and that they were given an overview of where their work in collecting data fitted into the overall policy issue being researched.
Despite our best efforts, the midwives collected only 25% of the data they were paid to collect. Managers and policy makers responded with naïve anger: “give me their names” said one. Rather than blame midwives who were feeling overworked and put-upon, we tried to understand what was happening sociologically, using the sociological concept of ideal types. In the article, we refer to different styles of adapting to demand of data collection over which they had little say or formal control. These terms were Repairers (the demands for data collection disrupted their work but they adapted both processes to try to make it work as best they could in difficult circumstances); Refractors (as often as possible, they would creatively use adherence to formal rules of their usual work procedures as a basis not to collect data) and Resistors (those who actively opposed the research but could/would not directly challenge their managers, and so used appeals to broader contexts to avoid collecting data). Managers did not challenge this evasion as they feared that midwives would move jobs and leave the service even more short-staffed than they already were. Although we paid research funds to cover the work of the midwives, these monies were not always passed on by the health organizations concerned, and this lack of financial transparency was also used by the midwives to justify their resistance.
Some sociologists have tended to reject all large scale research because of this “hired hands” problem. A current problem is that this model of research, in which ordinary nurses or midwives collect data for a project run by a medical researcher, is characteristic of much health services research. Put bluntly, much health services research would be of little value if it relies on data collected by workers who have no say over the process, no stake in the results, and who are denied any vision of where their efforts fit into an overall picture.
In our report to the commissioners, rather than either hide these problems (the reaction of many medical studies we feel) or rejecting results outright (the reaction of some schools of thought within sociology) we characterised the practical consequences for the quality of data collected. To take just one example: midwives sometimes did not recruit white mothers into the study (wrongly presuming that sickle cell only affects ethnic minorities, when in fact 1 in every 450 white babies born carries a gene relevant to sickle cell). Rather than ignore this or discount the research entirely as flawed, we tried to use data to estimate how often this occurred and to assess the impact of this on the overall findings and policy recommendations.
The lessons for the hired hand concept in sociology is that it requires refinement to take account of how work hierarchies interact with gender hierarchies and how contingencies of workplace organization may mean that some workers are required to be busier than others, so that extra work demands the impact upon them differently. The lesson for health services research is that it needs to take account of this phenomenon in the design of the study and in analysis of the results, but above all in the respect with which it treats health workers who collect data.