The Science Of Cooking: Understanding The Biolo...
CLICK HERE - https://urlin.us/2tmcMO
The Science Of Cooking: Understanding The Biolo...
Citation: Fonseca C, Pettitt J, Woollard A, Rutherford A, Bickmore W, Ferguson-Smith A, et al. (2023) People with more extreme attitudes towards science have self-confidence in their understanding of science, even if this is not justified. PLoS Biol 21(1): e3001915.
The model predicts that attitude strength should be positively correlated with subjective understanding and that this will be robust to covariate control (Prediction 1a). We consider age, religiosity, political identity, and educational attainment as covariates. Both educational attainment [1] and religiosity [18] are known to relate to science attitude. For more polarised sciences, but not typically for genetical issues such as GM, political identity can predict attitudinal position [1]. As age may well covary with both, and is likely to be relevant in the context of COVID, where the old were especially affected, we also include it as a covariate. Such a test need not establish that both extremes of attitudinal position score highly for subjective understanding. Thus, a second prediction of this model is that the same trends are seen as the distribution moves towards both extreme positive and extreme negative attitudes. We therefore ask whether the function relating attitudinal position, measured symmetrically, to subjective understanding is approximately U shaped. If so, the function should be better fitted by a quadratic model than by a linear model (Prediction 1b). However, an improved quadratic fit may yet be obtained if only one of the 2 attitudinal extremes is associated with increased subjective understanding, the other half of the data being uncorrelated. Assuming the quadratic fit is better, we additionally ask whether those with an attitudinal score greater than or equal to zero (neutral) show a positive correlation between subjective understanding and attitude that reverses to a negative correlation for those with an attitudinal score less than or equal to zero (Prediction 1c).
Several further differences between the GM attitude questions and the nonspecific Trust and Hype questions are notable. First, there are many more (approximately 5%) in the strong rejecting class than in the prior 2 strong rejector classes, these being 1% to 2% (S1A Fig). Second, this class has approximately 50:50 sex ratio, different to that for the other 2 (Fig 6). Third, for the GM question, education attainment is a much weaker predictor of attitudinal position and is indeed not significant on partial correlation analysis (Table 5, see also S1B Fig) nor on multitest correction. Further, unlike the prior 2 questions, age is a strong predictor (older people are more opposed to GM), before and after partial analysis (Table 5). Fourth, GM attitude is not correlated with political identity (as previously found [1]) while for both Trust and Hype, there is a tendency for those more oppositional to science to be more right wing (Table 2). Fourth, when we restrict analysis to cases where the attitudinal score is less than or equal to zero, we see no significant difference in the slopes relating knowledge score and subjective understanding to attitude for the GM issue. Fifth, many fewer of the 12 science knowledge questions are predictive of GM position than seen for the 2 more abstract questions and correlations are generally much weaker (mean rho for Trust = 0.094, Hype = 0.12, GM = 0.058: S1 Table).
These results come with numerous caveats, the most important of which question the generality of our results. For our survey data, we recover the classical positive correlation between knowledge and attitude (S1 Results) seen in cases where attitude to science is not affected by religion/political factors [3]. Whether our results extend to cases where high educational attainment is associated with both strongly positive and strongly negative attitudes [1,2] is unknown. Indeed, these instances tend to be those where religion/polit