| 1) Establishing the intention of the research |
As a research objective, the main objective of this study was adopted, as described in the introduction. |
| 2) Exploratory preliminary research with keywords in databases |
It consisted of choosing the keywords used as search terms, also serving as a check on their adherence to the research objective. Thus, the following words were tested: management and administration of agro-industrial cooperatives, social management, risk management in agribusiness, strategic management in agribusiness, new generation cooperatives, cooperative management, cooperative governance and challenges to cooperative system. Searches were performed in the leading academic content portals: CAPES Journal Portal, Web of Science, Scopus and Scientific Electronic Library Online (SciELO). |
| 3) Definition and combination of keywords and databases |
After verifying the adherence of the keywords, it was decided to keep the same terms used in the previous step, adding, however, variations for the plural and English and Spanish languages. Likewise, it was decided to concentrate the searches on the CAPES Journal Portal, considering that 95% of the references found in the other databases during the preliminary research were also present in the CAPES database. This also showed greater availability of complete references, greater agility and ease in searches. Due to classical references relevant to the themes addressed by this study, a temporal criterion was not adopted for the searches. |
| 4) Final search in the databases |
The last search led to the identification of 272 possible references. |
| 5) Filtering procedures |
The identified references were analyzed to eliminate duplicate results and out-of-context research, whose complete references are published in languages with complex translations, such as Chinese and Russian. After filtering, 124 possible references remained. |
| 6) Identification of impact factor, year of publication and number of citations |
As in Pagani et al. (2015Pagani, R. N., Kovaleski, J. L., & Resende, L. M. (2015). Methodi Ordinatio: a proposed methodology to select and rank relevant scientific papers encompassing the impact factor, number of citation, and year of publication. Scientometrics, 105(3), 2109-2135.), this step was developed simultaneously with number 08 since, when collecting the metrics, most texts were also found in full version. For this stage, the data collection sources were Google Scholar, the journal websites, Scimago Journal Rank (SJR) and Journal Citation Reports (JCR). Publications that do not have an impact factor, such as book chapters or papers from conference proceedings, received the value 0 (zero) for the impact factor. |
| 7) Ranking the papers using the InOrdinatio |
The methodology, developed initially by Pagani et al. (2015Pagani, R. N., Kovaleski, J. L., & Resende, L. M. (2015). Methodi Ordinatio: a proposed methodology to select and rank relevant scientific papers encompassing the impact factor, number of citation, and year of publication. Scientometrics, 105(3), 2109-2135.), aims to rank the selected publications in order of relevance and, therefore, uses the calculation of “InOrdinatio”. However, in the study developed here, the ranking was not intended to establish a cutoff point as a selection criterion for the analyzed papers to write the proposed literature review. Thus, the calculation performed was like that offered by Pagani et al. (2015Pagani, R. N., Kovaleski, J. L., & Resende, L. M. (2015). Methodi Ordinatio: a proposed methodology to select and rank relevant scientific papers encompassing the impact factor, number of citation, and year of publication. Scientometrics, 105(3), 2109-2135.). However, the relevance index (RI) was used, given by equation (1): |
|
|
| where IF is the impact factor, divided by 1000 (one thousand) to normalize it about the other criteria; α is the weight of the year of publication factor, which the researcher must assign, varying between 1 and 10, the higher the value, the more important the year of publication will be; APe is the year of the literature review; APu is the year of publication of a reference; and ∑Ci is the sum of citations of a reference. |
| To carry out this study, considering important classic references for the topics covered, α = 5 was established. In this way, it was possible to standardize the weight of the criteria, year of publication and number of citations. On the other hand, Pagani et al. (2015Pagani, R. N., Kovaleski, J. L., & Resende, L. M. (2015). Methodi Ordinatio: a proposed methodology to select and rank relevant scientific papers encompassing the impact factor, number of citation, and year of publication. Scientometrics, 105(3), 2109-2135.) assert that when temporal filters longer than ten years are adopted, or as is the case in this study, temporal filters are not adopted, it is common for the RI to present negative values. Thus, aiming at the scope of the references that were finally selected, the cutoff point of RI < -80 was adopted here. Thus, applying equation (1) to the 124 references remaining after the filtering stage, there were 89 references considered relevant. |
| 8) Finding the full papers |
Most of this stage was developed along with stage 6. Only seven references were not found in the full version before, acquired after more detailed searches. |
| 9) Final reading and systematic analysis of the papers |
At this stage, the 86 references considered relevant were read, analyzed and registered, gathering the necessary conditions for the development of this study. |
