One thing I found really interesting about this research is statements like these:
Therefore, though transformational leadership has been contrasted to transactional leadership (with the former being suggested to be superior), the use of contingent reward behaviours seems similarly effective to transformational leadership.
It sounds very believable to me that ~0% of ânonobviousâ leadership recommendations donât outperform a âplaceboâ. (Or, as you suggest, are only good subject to contingencies like personal fit.)
I would be curious if doing this review gave you a sense of what the âcontrol groupâ for leadership could be?
Iâm imagining something like:
Your team has reasonably well defined goals
Your team has the ability to make progress towards those goals
Your team is not distracted from those goals by some major problem (e.g. morale, bureaucracy)
We might hypothesize that any team which meets 1-3 will not have its performance improved by âtransformationalâ leadership etc.
Do you know if anyone has studied or hypothesized such a thing? If not, do you have a sense from your research of what this might look like?
<<Do you know if anyone has studied or hypothesized such a thing?>>
No, but my research was very brief and focused mostly on reviews and meta-analyses rather than looking at the primary studies. So a more thorough research project (or interviews with the authors of the reviews and meta-analyses?) might reveal this sort of thing better.
<<If not, do you have a sense from your research of what this might look like?>>
Iâm not sure Iâm quite following, but if you are asking what the control group looks like currently:
Usually the research is just correlational. So implicitly, itâs comparing high scores to low scores on the same scales.
Some training evaluations have used experimental or quasi-experimental designs, randomising participants to receive training or receive no training, I believe.
If youâre asking if I can imagine there being more targeted sorts of research that address your specific hypothesis (âany team which meets 1-3 will not have its performance improved by âtransformationalâ leadership etcâ), I can certainly imagine it. E.g. you run a similar experiment across lots of different organisations, and beforehand, you assign the organisation some sort of score (subjective rankings out of 10?) for each of those variables and see whether there are correlations. But 1) this sounds like a very intensive research programme, 2) I donât usually see social scientists and academics use subjective rankings as variables, perhaps because it seems less replicable by other researchers and less rigorous?
One thing I found really interesting about this research is statements like these:
It sounds very believable to me that ~0% of ânonobviousâ leadership recommendations donât outperform a âplaceboâ. (Or, as you suggest, are only good subject to contingencies like personal fit.)
I would be curious if doing this review gave you a sense of what the âcontrol groupâ for leadership could be?
Iâm imagining something like:
Your team has reasonably well defined goals
Your team has the ability to make progress towards those goals
Your team is not distracted from those goals by some major problem (e.g. morale, bureaucracy)
We might hypothesize that any team which meets 1-3 will not have its performance improved by âtransformationalâ leadership etc.
Do you know if anyone has studied or hypothesized such a thing? If not, do you have a sense from your research of what this might look like?
<<Do you know if anyone has studied or hypothesized such a thing?>>
No, but my research was very brief and focused mostly on reviews and meta-analyses rather than looking at the primary studies. So a more thorough research project (or interviews with the authors of the reviews and meta-analyses?) might reveal this sort of thing better.
<<If not, do you have a sense from your research of what this might look like?>>
Iâm not sure Iâm quite following, but if you are asking what the control group looks like currently:
Usually the research is just correlational. So implicitly, itâs comparing high scores to low scores on the same scales.
Some training evaluations have used experimental or quasi-experimental designs, randomising participants to receive training or receive no training, I believe.
If youâre asking if I can imagine there being more targeted sorts of research that address your specific hypothesis (âany team which meets 1-3 will not have its performance improved by âtransformationalâ leadership etcâ), I can certainly imagine it. E.g. you run a similar experiment across lots of different organisations, and beforehand, you assign the organisation some sort of score (subjective rankings out of 10?) for each of those variables and see whether there are correlations. But 1) this sounds like a very intensive research programme, 2) I donât usually see social scientists and academics use subjective rankings as variables, perhaps because it seems less replicable by other researchers and less rigorous?