Given the growing complexity dynamism and instability of contemporary project analysis, all businesses and economic actors including patient and public involvement should to be able to ascertain factors or variables that might influence its relationships with the project or market for sustainable and reasonable inferences. though this approach may be expensive, it is the only way one can unravel the complexity involved and ensure that stakeholders take ownership of the solution which ensures adoption and implementation. It is the researcher's responsibility to justify their involvement and be able to explain to those to be involve their role and how they will benefit from such projects

There are many ways that PPI could help with secondary data analysis - but as with every question about PPI the answer is always - it depends - because context is key. So some of the ways it could help include:

- helping define / refine the questions you want to ask

- identifying questions that haven't been asked or gaps in the data/ literature

- reflecting on initial themes/ conclusions that researchers draw out

- thinking about who needs to hear about the conclusions, what to tell them and how to tell them

The extent of the impact of involvement is likely to depend on the topic being addressed and nature of the analysis. For example, patients are more likely to contribute to qualitative analysis of interview data than statistical analysis of clinical data. 

None of the tasks listed above require specialist skills on behalf of the patients you involve - because you are simply asking them to reflect on their experience and how it relates to your work. However, as researchers you might need to think about how to enable the people involved to do this task - I could point you in the direction of a number of published case studies where researchers have done this. 

PPI in data analysis is not necessarily about having patients alongside you doing the same job as the researchers. Though this has been done - and the service user researchers tend to draw out different themes from their analysis - this has also been documented and reported. In this case - yes the patients/ service users do need to be trained as researchers - but this may not always be necessary.

To answer your question in more detail I'd really need to know the specifics of what you're working on and why you want to involve patients/ members of the public - then how it could work for you in your specific context would be easier to comment on.

Hope this is helpful.