I am really struggling to understand what you are attempting to measure. Recruitment and selection as dependent variables are such huge topics. I can think of a dozen variables associated with both recruitment and selection by themselves.

Can you explain in greater detail what you are trying to accomplish? The goal of recruitment and selection is to attract, identify, and bring into an organization employees who enhance the organization's ability to add value. Do you wish to measure the increment of that value added? How do you plan to measure the quality of persons ultimately brought into an organization and how will you accurately assess whether the independent variable (the degree of centralization or decentralization) actually contributed to the incremental difference?

It seems to me that you have a very complex research problem to attempt to set up and accurately measure. Perhaps I just do not understand what you intend to do . . . . but I find your question intriguing because of its inherent complexity.

CC

centralization can be measured through authority concentration e.g an organisation with hierarchical structure has centralization of authority, whereas in flatter organisation structure there is decentralization of authority. In HR practices if team approach is adopted authority gets distributed but in case of one man approach authority is centralized.

Regards

Since you are studying government organizations, there is need to check organization structure – central HR, division/ department HR and local office. Further analyse involvement of line managers in HR activities.

At what level policies are decided and who executes operational part of HR. The options could be

· Central HR, division/ department HR and local office

· HR Only, HR in consultation with line management, line management with consultation of HR, only line management

(Reference – Policy and practice in European Human Resource Management – The Price Waterhouse Cranfield Survey)

Hope this can help to analyse centralization/ decentralization.

I think each area had problem and charakteristik so not same centralization and desentralization

Social network anysis and Organziation network analysis. Look for the work conducted by Professor Martine R. Hass from the Univ. of Pensylvania on people analytics and collaboration.

At a practical level, if the organisation(s) you are studying have a central set of Job Descriptions/Competency Frameworks, or even a Job Evaluation handbook with factors for each grade/band, you could compare these to see what is the level of devolved authority for each role. You could cross reference them all first, to get a benchmark for what their terminology means. The Chartered Institute of Personnel & Development may have further information on what your benchmark might be: https://www.cipd.co.uk/

Thank you all for your answers. To clarify my question I would add that my main intention first is to measure the level of centralisation of HR practices. In order to achieve this I am conducting interviews in the ministres on level of involvement of different structures (HR,line management,central HT ect) in recruitment and selection processes of state employees. I think it would help me to find out whether HR practices in particular countries are centralised and decentralised. The next step would be to get an opinion of state employees and responsible parties on effect of those two practices. I am agree with Mr. Cam Caldwell that recruitment and selection as dependent variable is a huge topics.

But , for now I am not intending to get into the value of the persons are recruited. I am trying to identify which system is preferable to avoid political patronage,nepotism during the recruitment and selection processes and enable meritocracy.

According to Prof. David Ulrich from www.rbl.net you can decentralized, centralized and mixed models. Key point is that they folow the business needs. Notwithstanding, governments tend to be centralized. So I guess understanding the underlying reasons behind that is the key: good work!