Clarify that there has been a cleaning process.

[Nikoleta-v3]

Second, I am wondering whether you had conducted manual screening of collected literature to ensure their close relation to the research target. Different from systematic review aiming to provide in-depth analysis of representative literature, bibliometric analysis is used for big data analysis to enable a comprehensive and general view. Thus, sufficient coverage of relevant literature is especially essential for a bibliometric study. In particular, you included articles which had used any of the search terms in the title, the abstract or the text. The use of title and abstract is understandable since they are commonly used to search articles (you can refer to bibliometric studies such as https://www.sciencedirect.com/science/article/pii/S0360131520300555, https://bera-journals.onlinelibrary.wiley.com/doi/full/10.1111/bjet.12907?casa_token=vCNbiEvDF5gAAAAA%3Ap9AeWKgtwePj-ABgaPIFecQrTObZdjh26sG49aeHs47Ms7zHROw8I_JhUpvQnOlZ0Vx5oCrzwmCN1kTsVQ, https://idp.springer.com/authorize/casa?redirect_uri=https://link.springer.com/article/10.1007/s11042-020-09062-7&casa_token=z8E4b9PYoncAAAAA:9X2oCiTpIRj2V6xMkkLsjXvAyD2NSmMxVNYK_K-fLdb67SLc3zUFxoCmmniFJCNprc20XInB9KPum71LMCw ), however, the inclusion of full text can be troublesome as it may result in too many irrelevant articles being included. Thus, manual screening is needed and should be conducted to ensure data quality.

The data set has been cleaned. Paper's that did not fit the narrative were removed. However, there might still a few articles that do not?

[Nikoleta-v3]

Reviewer 2

I was not convinced that the dataset covered all relevant article and only the relevant articles. It was not described at all what steps were taken to find the grey literature that was not covered by the database searches. Did you hand-searched the citation of eligible studies or consulted with experts? These are all very important steps to get a comprehensive review. I know that some steps were taken as the author mentions to include some extra paper. Further, what steps were taken to ensure that only relevant articles were included? An abstract can mention Prisoner’s Dilemma but not analyze this topic. Usually, it is good to scope the abstract (at least) to check whether studies were of interest. Further, the number of duplicates should be reported.

[Nikoleta-v3]

The flow chart will help with explaining the above better.

[Nikoleta-v3]

Possibly include a random sample (n=20) to approximate the mistakes in the data set.