Critical Analysis -Analyzing ‘Bad’ Science

Introduction

In the article “Where There Are Girls, There Are Cats” written by Li and colleagues, the focus is directed towards understanding the way free-ranging cats manage to live with humans. The study targeted 30 Universities in China, and it has claimed that there is a linear proportion between free-ranging cats and female students within the targeted Universities. The study showed that female students were increasingly concerned about the well-being of free-ranging cats and that the population density of free-ranging cats is dependent on the number of female students within these Universities. The researchers concluded that human gender plays a significant role in understanding how well the population of free-ranging cats can be controlled, based on the concern towards free-ranging cats between males and females. Therefore, in line with these claims, this paper will conduct a critical analysis of this research to understand the correlation and causation and some of the mistakes evident from this research.

Background

This article has so far been removed and is no longer published. Peer review plays a significant role in scientific processes.  According to Science Direct, which is a leading platform for peer-reviewed literature, an article is removed from its database due to numerous reasons. These include articles that contain errors, accidentally submitted articles, articles that present infringements of profession’s ethical codes, and, most important, articles that contain bogus claims of a particular study (Li et al., 2020). Furthermore, scientists appraise and critique studies, and research that fails to undergo this process is not as reputable and can be flawed (Brunning, 2014). Considering that this article is research undertaken by experts from the school of Life Science And Department of Biological Sciences and Technology in the respective institution, the article can be classified as a peer-reviewed article. However, experts have raised man issues concerning this research, forcing it to be removed from journal databases. It can be pre-assumed that this research does not meet the set peer-reviewed quality standards. Additionally, the study is flawed because perhaps the researchers might have failed to pass it through other experts in this field to appraise and offer critiques before having it published.

Methodology

The study targeted female and male participants from colleges with the Yangtze River Delta in China. In this region, the researchers identified a total of 54 colleges and selected a sample of 30 out of this population. Participants in this study were chosen randomly, where 2066 questionnaires were administered through an online platform. Out of these, only 2033 were accepted back after the exclusion of 33 questionnaires that were wrongly filed. In this view, this study is likely to have unreplaceable results

In this study, there is evidence of blind testing, which alludes to conducting tests on participants who are unaware of whether or not they are in the experimental or control group. The collected data from numerous trials one was through 2033 questionnaires and the socialization test of 27 participants. It is not clear whether the participants took part in the entire test or not. Moreover, data on the number of students from each University cannot be considered accurate, given that it was obtained from the school’s website (Li et al., 2020). In other words, the results do not seem to be considering that they are not tested over a wide range of conditions to make sure that they are consistent (Brunning, 2014). There is no explanation on how a correct number of students were arrived at considering that such data is not accurately and readily available from the Universities websites. Another critical issue is on how the number of free-ranging cats was reached because the researchers calculated only the percentage of greenery coverage of each campus through a high-resolution aerial photograph other than observing the number of cats to minimize the error.

The information provided is not statistically sound. The participants who gave their responses in the questionnaires were expected to come from the institutions within the Yangtze River Delta. However, the researchers state that the 33 questionnaires that were excluded from the analysis were those that were filled with people outside of China and those with wrong answers. This means that they included answers that came from outside their area of study, and this might have impacted their results and conclusions.

            In general, the methodical ambiguities are based on the fact that the researchers took the socialization tests that suggested that cats were responding better to females than males by involving only 16 students, 7 females and 9 males.  Again, it’s only data from interaction with 27 cats that were analyzed.  Even without considering the sample size, it is not clear whether the 27 cats had contact with the entire sample of 16 students, and this may not look like a significant distinction between a low sample size, which is useful if represented as a qualitative study.

Interpretation of results

As a way of testing the hypothesis, the researchers have used the data analysis, questionnaires, and socialization techniques or tests. In the first test, the researchers have employed the correlation between the free-ranging cats’ population together with the human population from 30 Universities. The research has associated the cats’ density with female students’ portion, explains why the correlation exists between these two, and what could be the impacts with female students. From the questionnaires, it was identified that more females had either fed or rescued a free-ranging cat as compared to males. Finally, in the socialization test that comprised of 16 participants, 9 were males and 7 females who took part in the study of a host of 27 free-ranging cats (Li et al., 2020). The researchers offer a generalized conclusion that in the socialization test, it was clear that cats might potentially differentiate human males from human females. Here all the methods used in data collection have been tested; all the data has been considered, and numerous models such as the general linear model have also been utilized.

Could the data be falsified? How?

Falsification of data refers to the omission or changing of the research results to offer support to the claims, hypothesis, or even other data. It can also include manipulation of the research instruments procedures, and materials. In general, the falsifiability guarantees that if the claim is false, the evidence provided from data collected proves it false, and in case it’s a valid claim, the evidence does not disapprove it (Lett, 2007).  Falsification can be evident in images manipulated in a way that distorts the data. In the research Where There Are Girls, There Are Cats falsification is not directly on the data collected, but on the way the data is interpreted. However, it’s crucial to note that the claim in this research can be accepted tentatively as true until evidence is brought forward to disapprove it. Having that in mind, the study is dwelling more on estimates and assumptions. There is an obvious bias on how the number of the cats’ population in all targeted Universities was selected in terms of inclusion and exclusion. There is also an increased generation of data; For instance, data collected from all the 30 Universities was done in a sample of them and assumed that the population density of free-ranging cats in all the campuses could not have notable disparities. Therefore, as much as the researchers tried to minimize bias, the population of cats in various Universities could have been affected by other aspects such as availability of food and access to people who could feed them. Therefore, generalizing the number of cats found on each campus could have affected the final result.

The data collected in this research is sufficient, but its reliability that measures the consistency of results, the precision, repeatability, and trustworthiness is doubtful. As noted, there are numerous errors evident from the way the researchers generalized most of the data collected. This means that there is no consistency in measurement across time and multiple instruments. For instance, the researcher did not report on the measures taken to ensure that other people from Universities outside the Yangtze River Delta did not respond to the questionnaires (Li et al., 2020). This implies that the observed score of measures does not reflect the correct score of the results.

The interpretation of the results is not based on Logic and not in line with what the results say.  Additionally, Logic implies that any argument that the researchers present as evidence support to the claim have to be sound. Conclusions are drawn from the analysis of other studies other than the data collected and analyzed. For example, the researcher states that, under the powerful selective pressure from humans, cats perhaps learned to socialize while communicating with human females to get more food. In this regard, this argument does not deem valid because the conclusion follows unavoidably from its premises (Lett, 2007). The socialization test reveals that cats were more sociable to human females than human males, especially under the condition of bad weather. Again, the research makes another assumption that during the sunny days, the socialization action of cats with humans is because they have enough food. This is a general assumption that is not backed with evidence from the data that was initially collected. Here, the implication is that it’s not that cats socialize most within humans, but other factors such as food availability and weather conditions play a crucial role.

The correlation does not equal causation, or there are false connections given that most of the conclusions made are based on assumptions other than the data collected. For instance, bad weather does not mean that cats will be forced to socialize mostly with males than females. Moreover, the issue of comprehensiveness is in doubt, considering that the evidence provided to offer support to the claim does not seem exhaustive, meaning that all the available evidence has not been considered (Lett, 2007). For instance, there are other factors that could result in this socialization behavior, such as food availability and scarcity during certain weather conditions, which could cause socialization to increase. For palpable reasons, it is not ever reasonable to put into consideration only the evidence that backs a theory and to abandon the evidence that is in contradiction with it (Lett, 2007). The data has been misinterpreted because, as stated, most of the conclusions made are based on assumptions that are not backed with facts from the data collected.

Conclusions

The study has unsupported conclusions supported by evidence of oversimplified findings, for instance, the data relating to cat density in relation to Greenery coverage in each campus, density in the proportion of female students, and density in feeding. It’s important to note that any research must be clear on the facts that researchers are trying to prove and the conclusions that are not supported (Brunning, 2014). In other words, the series of the actual causes for socialization between cats and humans have been reduced, and there is no longer a genuine causal connection between the supposed causes and the definite effect. This means that the numerous reasons for the socialization revolve around one assumption that females are better and can assist in the management of free-ranging cats’ population. The results being analyzed are based on the data collected; however, the discussion and conclusions made from the data are generalized.  For example, the results are based on whether Chinese human females have increased concern over free-ranging cats as compared to human males. The conclusion made is that more females’ feed free-ranging cats as compared to males. In the case of socialization, climatic factors have been factored in and identified to affect the way cats react to both females and males. When it comes to the interpretation of results, the generalization of factors comes in and influences how results are interpreted. That is in terms of correlation and causation, and it does not mean that because many human females are likely to take care of free-ranging cats more than to human males, free-ranging cats will tend to be attracted to human females than males (Brunning, 2014).  Additionally, the researchers state that it’s hard to determine the cat’s cause of preference for females over males in terms of behavior. This implies that it’s not the females’ concern over free ganging cats that makes these creatures to prefer them over males because factors that include how human females react towards these cats can also be different. The researchers seem to be forcing results and conclusions by indicating that it is cats that attract females without explicit support from the result interpretations. In general, the conclusion is supported by the data, but many questions require to be answered, which cannot be obtained from the current study. In other words, the research data, results, interpretations, and inferences drawn are full of obvious errors. It is not clear whether the researchers collected the data about the number of free-ranging cats, all 30 institutions, because, as noted, data were sampled from a few Universities and then generalized.

Conclusion

In general, the research was undertaken with all measures put in place, and the data collection methods, interpretation, and discussion is clear. However, the major issue with this research is the generation and making findings based on assumptions. As shown in the first most crucial rules of FiLCHeRS, which are falsifiability, Logic, and comprehensiveness, the conclusions made are not entirely based on the evidence from the data collected. Moreover, the correlation on how cats’ socialization connects with human females is not clear. In contrast, the data collected on the free-ranging cats was assumed to be the same in 30 Universities.

References

Brunning, A. (2014). A Rough Guide to Spotting Bad Science. Compound Interest Blog.

Lett, J. (2007). A field guide to critical thinking.

Li, Y., Wan, Y., Zhang, Y., Gong, Z., & Li, Z. (2020). Temporary Removal: Where there are girls, there are cats. Biological Conservation, 108412.

Li, Y., Wan, Y., Zhang, Y., Gong, Z., & Li, Z. Where there are girls, there are cats.