Mid-term evaluation is performed during the second stage of the programming cycle, during the implementation of the interventions.
Depending on the conclusions of mid-term evaluation, adjustments may be made during the cycle. This evaluation critically analyses the first outputs and results of interventions. It also assesses the financial management of the programme and the quality of the monitoring and of its implementation. It shows how and whether original intentions have been carried out and, where relevant, checks whether de facto changes have been made to the initial objectives. By comparison with the initial situation, it highlights changes in the general economic and social context and judges whether the objectives remain relevant. Mid-term evaluation also examines whether the evolution of Community priorities and policies poses a problem of coherence, and helps to prepare adjustments and reprogramming, and to argue them in a transparent manner.
Mid-term evaluation relies heavily on information drawn from the monitoring system, but also on ex ante evaluation and on information on the context and its evolution. It generally consists of short and exhaustive exercises focusing primarily on the results of the programme evaluated, without attempting an in-depth analysis of impacts that have not yet had the time to emerge. It is, however, possible and advisable to refer to in-depth or thematic evaluations of former programmes when such analyses do exist. Mid-term evaluation has a "formative" nature, that is to say, it produces direct feedback into the programme that it is helping to improve as far as its management is concerned.
For the 2007-2013 period, mid term evaluation has been replaced by ongoing evaluation, preferably based on a multi-annual evaluation plan. This new approach aims to overcome the rigidities of the 2003 mid term evaluation and facilitate the evaluation of parts or themes across programmes at a time when those responsible for the programme have a need for information, evidence, analysis or judgements from evaluators.
An evaluation plan should, ideally, outline responsibilities for evaluation, the budget, an indicative list of evaluations to be carried out (which can be reviewed and updated periodically), the mechanisms for designing and managing evaluations (tendering, steering groups, etc.), and arrangements for using evaluations (dissemination, responding to recommendations, monitoring the implementation of agreed recommendations, etc.). Evaluations plans can be established nationally or at the level of an individual programme (either regional or sectoral).
The concept of ongoing evaluation is extremely flexible. By leaving it entirely to the discretion of the programme authorities to decide what should be evaluated and when, evaluation should become more of a management tool to help the performance of the programme. Programme authorities can, of course, decide to undertake mid term evaluations of entire programmes, as outlined above. However, given that entire programmes need no longer to be evaluated at fixed points in time, there is also an opportunity to carry out in-depth evaluations of particular areas of intervention, perhaps using more complex and rigorous methods. An ongoing programme of evaluation should support public authorities to find out "what works" in their programmes and what doesn't and to take remedial action where necessary.
In the 2007-2013 period, the only requirement for evaluation during the programming period is when a modification to the programme is made because of a significant departure from the objectives originally set. Given the time required for launching and carrying out evaluations, this emphasises the need for evaluation to be planned pro-actively, to ensure that evaluation evidence is available when it is needed.