Author: Raidell Avello Martínez – Translation: Erika-Lucia Gonzalez-Carrion

Expressing the limitations of the study is a task that many times, especially by novice researchers, tend to be avoided. Many think that the limitations detract from the research carried out over a long time and a lot of effort. However, it is quite the opposite, expressing the limitations of a study gives more validity and rigor to the research process developed.

When the authors explain and comment on the limitations of their study, they show greater mastery of the characteristics of the population or phenomenon evaluated, the methodology and instruments applied, the scope of the results obtained and the theoretical and investigative body that are part of the antecedents of The research carried out, that is, that far from demerit the findings obtained gives them added value of rigor and validity.

In the American Psychological Association’s own publication manual (APA, p. 36), it is suggested to recognize the limitations of your research, and to point out alternative explanations of the results; discuss the real possibilities of generalization and external validity of the results. In addition, it should be noted that this analysis should take into account the differences between the target populations and the sample involved in the research; And for interventions, discuss the characteristics that may make them more or less applicable in circumstances not taken into account in the study, comment on the measurements used and which could have been used, the measurement time, incentives, compliance rates, among others Elements that reveal the importance of expressing the limitations.

Despite the importance of this section in scientific articles, many papers are published without these elements. Responsibility for this phenomenon rests primarily with the authors, but I believe it is the responsibility of the reviewers to promote and demand that the limitations of the work they review are evident.

Some fragments of limitations in articles published in academic journals (especially in the field of educational technology) can illustrate these comments:

• IRRODL, 2014, 15(6) p. 59, “Whilst the goal of phenomenographic research is to achieve a complete picture of variation, the results can never be completely representative of all the different ways of conceptualizing or experiencing a phenomenon (Åkerlind, 2005). Because of the scope of this initial research project, the variation represented in the results is limited in accordance with the number of participants. Generally, it is recommended that phenomenographic studies involve 15 to 20 participants (Trigwell, 2000). This preliminary study involved only five. Additional studies should be done in order to more fully examine the variation of experience. Finally, the results may not necessarily be generalized to other programs or institutions as the participants are all members of the same master’s program, which may, itself, be idiosyncratic”.

• Pixel-bit, 2014, 45, p. 33, “The number of participants hospitalized is lower than initially estimated. The main reason was the reduced number of hospitalized students who met the pre-requisites for collecting information. It is very difficult to make predictions about the number of students who can participate during a period, it is for this reason that the duration of data collection should be very broad, so that the number of participants could increase. Regarding the design of the ALTER tool, there was a lack of previous consensus with the teachers, since, although the interest of the teachers in participating was evident, they did not want to invest as much time as would have been the ideal one, reason why we infer that our vision on their participation may have been erroneous. “

• Comunicar, 2015, 45, p.35, “Among the limitations of the research is that although the methodology achieves the main objectives, enabling a deep and direct explanation of the Internet as a source of opportunities for active aging, some issues of interest would require a Complementary treatment, to be considered as a first approach that requires greater depth “.

• Educational Technology & Society, 2016, 19 (4), p.84, “This study may have been limited by its structure. First, the experiment was conducted online in the form of a college assignment over 1 week, and the learners participated individually. Although the instructor provided special instructions in class for carrying out the assignment, the students’ multimedia learning may have been interrupted. Nevertheless, the findings of the study provide insight into the effectiveness of various types of WOE used in pre-training to optimize the cognitive load and enhance learners’ comprehension of the content. Second, learning can be affected by various elements such as synchronization, pauses, and interactions that were not considered in this study. It seems necessary to explore the effects of interactions in further studies to lead to successful learning experiences”.

• RED, 2016, 52, Artíc. 5, p.22, “Regarding the validation of the instrument, despite some limitations, satisfactory results have been obtained in both convergent validity tests. Likewise, despite the discriminant validity problems detected between behavioral intent and utility, the factorial analysis carried out reflects an acceptable goodness of fit model. “

As can be seen in these fragments, some more precise than others in highlighting the limitations, somehow reflect the main deficiencies of their results to be taken into account in possible replications of these investigations, which is very useful to the scientific community in The difficult task of “not to stumble again with the same stone”.