The Black Pedagogy Scale: A New Task to Explore Educational Practices for Children’s Well-Being

The present contribute focuses on the concept of “Black Pedagogy” (Rutschky, 1977; ISBN: 3548356702), meant as a set of educational practices assimilable into those that nowadays are included in the frame of physical and psychological maltreatment (e.g., corporal punishment, frightening children, etc.). The purpose of this work is to present our operationalization proposal of the concept and the results deriving from a first validation of the “Black Pedagogy Scale”. The questionnaire was administered to 374 Italian university students in their university classrooms (pilot study with double administration) and to 830 Italian adults, parents of primary school-aged children, through an online survey platform (main study). In the pilot study, explorative analyses, paired-samples t-test and ML EFA (with Varimax rotation) were performed. In the main study, proprieties of the refined instrument and relations between the construct of Black Pedagogy and demographics were explored. The Black Pedagogy Scale (α > .8) resulted composed by three factors, consistently with what was initially hypothesized: “Values of Black Pedagogy” (var. 18.7%), “Education of children over time” (var. 10.6%) “Methods of Black Pedagogy” (var. 8.6%). Participants resulted more in agreement with Black Pedagogy’s values rather than with its methods, and those with higher educational qualification showed less agreement with the construct, F(2, 813) = 28.22, p < .001, η² = .065. The possible legacy of a Black Pedagogy’s forma mentis can contribute to explain why some detrimental disciplinary practices are culturally deemed as acceptable. Results suggest designing interventions focused on educational values to discourage such practices.

as well as denying emotional responsiveness, exploiting, rejecting and ignoring. Corporal punishment is instead referred to "any punishment in which physical force is used and intended to cause some degree of pain or discomfort, however light. It mostly involves hitting ('smacking', 'slapping', 'spanking') children, with the hand or with an implement" (UN Committee on the Rights of the Child, 2011, paragraph 24). Physical violence for disciplinary purposes (e.g., harsh treatment, and cruel or humiliating punishment) is still common in the context of families both in industrialized and in developing countries (Durrant, 2005;Pinheiro, 2006), and it often coexists with psychological forms of violence harmful to children's well-being both in domestic and in school contexts (Pinheiro, 2006). According to Perticari (2016), who edited the Italian translation of Katharina Rutschky's Schwarze Pädagogik, this level of children abuse is a contemporary problem and it deserves urgent attention, since this pathogenic education built on a devious authoritarian mentality is infecting children's everyday life and disguising abuse as a form of love and care. The author specifies that such level of maltreatment is covert, difficult to recognize and also very complicated: most adults do not realize they behave in a harmful way when slapping, yelling at kids or humiliating them, and they are instead convinced that they are acting in children's best interests. In fact, such disciplinary measures are considered necessary to promote a healthy and robust upbringing and, consequently, adults tend to justify or minimize them, often not recognizing them as something from which children have to be protected.
In the face of such considerations, a reference to the Italian legislation on the subject is necessary: before the reform of Family Law occurred in Italy in 1975, the abrogated article 319 of Italian Civil Code (I.C.C.) explicitly acknowledged the parent's power of restraining a child's misconduct, and it was interpreted as a sort of exemption from responsibility for the harmful acts committed by the parent towards the child if they were aimed at repressing bad behaviors, this being an area of exemption connected to the ius corrigendi (i.e., the right to correct), a corollary of parental potestas (Paladini, 2012). The existence of ius corrigendi is deduced from Art. 571 of Italian Penal Code (I.P.C.) that defines the offense of "Abuse of means of correction or discipline". The fact of referring to an abuse implies that there is a legitimate and permitted use of disciplinary measures, which can result in abuse if the measure is excessive, arbitrary or untimely (Ferraro, 2008). Moreover, the literature reports that with specific reference to Italian family relationships, part of the doctrine considers vis modica (i.e., moderate violence) a licit means of correction and also that it would be difficult to imagine the prospect of completely banishing it from family context (Catullo, 2012;Tortorelli, 2014). It seems therefore important to ponder further on two aspects: firstly, exploring whether and how the distinction between licit and illicit means of correction may be clear in the minds of educators -both parents and teachers -. Secondly, verifying if and how they can realize they are mistaken when applying such educational practices (e.g., a slap or a verbal insult) that they have seen widely adopted by the previous generations (parents, grandparents, etc.) without questioning their legitimacy. These two aspects are based on the hypothesis that the ongoing practice of the subtlest forms of disciplinary physical or mental violence coincides with a persistence in our society of the hierarchical and authoritarian model of the family mentioned above, which appears to be well described by Black Pedagogy values. Therefore, a measuring instrument capable of grasping such authoritarian educational model was needed. At first, we considered using the already existing "Poisonous Pedagogy Scale" within the O'Brien Multiphasic Narcissism Inventory (O'Brien, 1987), a subscale elaborated on the basis of Alice Miller's definition of Black Pedagogy. It measures the belief of having the ability to control others by taking advantage of one's own superordinate position (Montebarocci et al., 2003;Sines, Waller, Meyer, & Wigley, 2008). Nevertheless, we decided not to use such a scale and to develop a new one, in order to include more detailed and varied aspects of the Black Pedagogy concept, in line with the research interests of the present study. Florio,Caso,& Castelli 333

Purposes of the Study
This work is part of Doctoral thesis (Florio, 2018) (Chiorri, 2011). Two main sections composed the instrument in its first version: "Black Pedagogy Observation" (BPO hereinafter) and "Estimations of Black Pedagogy Diffusion" (ED). The BPO explores the construct of Black Pedagogy as it has been described in the literature, with 41 items (e.g., "Bad habits and character flaws must be eliminated through education", "Children should be kept constantly under control", and "Children must learn to be humble") and a response set designed as an agreement 4-point Likert scale.
Therefore, participants were asked to take a clear stance for or against each statement, and a possible change in the response scale would have been evaluated after analyzing missing data patterns. The ED section instead, gathers the estimations of the current diffusion of Black Pedagogy practices on our territory, and of the diffusion of these practices in the past. Therefore, a forty-second item (i.e., "By educational means 'used in the past' are meant the educational practices that took place in Italy from the post-World War II period until the 1980s") was added, in order to assess the time period to which participants referred when thinking of "the past", namely when such practices were widespread. This second section consists of two identical 12-items lists of the disciplinary practices typical of Black Pedagogy educational style (e.g., "Pedagogical beating (slaps, caning, etc.)" and "Treating the child coldly as a consequence of his/her disobedience"): participants have to evaluate the diffusion of each method one first time in respect to the past, and a second time in respect to the present days. Items of the second section were also accompanied by a 4-point Likert scale response set, but based on frequency instead of agreement. Clear instructions were given in order to inform respondents of the change in response options meanings (1 = "Not present at all"; 2 = "Present, but not common"; 3 = "Present"; 4 = "Widespread").

Participants and Procedures
Participants of the pilot study were Italian university students of the Department of Human and Social Sciences In the main study, the final version of Black Pedagogy Scale has been administered through an online survey platform to a sample of 830 adult subjects, recruited in the context of Northern Italy primary schools with the support and consent of school Head Teachers. The following table (Table 2)

Results of the Pilot Study
For what concerns the pilot study, a first exploration of the proprieties of 41-items BPO section was conducted in order to investigate its reliability and the distribution of responses. Table 3 presents a summary of the exploration of BPO section (minimum score 41 -maximum score 164) at Time 1 and at Time 2. The mean was similar to the 5% Trimmed Mean, thus indicating that it was not necessary to exclude outliers. In terms of mean response, the result was a value of 2.5 both at Time 1 and at Time 2 (minimum response 1maximum response 4). The Little's MCAR Test was not significant (p > .05) thus suggesting that data were missing completely at random. Cronbach's α resulted in an adequate value of .83 at Time 1 and .82 at Time 2, suggesting reliability of the scale (DeVellis, 2016). The distribution is slightly heavy-tailed (Westfall, 2014) and characterized by a positive asymmetry towards lower values of the scale in both occasions.
At Time2, both Kolmogorov-Smirnov and Shapiro-Wilk normality test resulted significant (p < .05) thus suggesting violation of normality assumption. Nevertheless, values of skewness and kurtosis included between a range The Black Pedagogy Scale 336 of -1 and +1 are considered acceptable (Muthén & Kaplan, 1985) and the histogram, as well as the Normal Q-Q Plot, showed that responses were reasonably normally distributed. The condition of our data, i.e. violating normality but reasonably normally distributed, is commonly encountered in social sciences, and Maximum Likelihood (ML) approach has been chosen since it is still recommended when a sever violation of normality is not present (Costello & Osborne, 2005;Fabrigar, MacCallum, Wegener, & Strahan, 1999). Data resulted adequate for factor analysis since Kaiser-Meyer-Olkin value was .79 (Kaiser, 1970(Kaiser, , 1974 and Bartlett's (1954) Test of Sphericity was significant (p < .001). Subsequently, ML Exploratory Factor Analysis (EFA) with Varimax rotation has been performed. After the first output of EFA it was decided to force the extraction of three factors following Kaiser's criterion of eigenvalues > 1 (Kaiser, 1960) and on the basis of the inspection of the Scree Plot, which clearly showed a change of direction after the fourth dot (Cattell, 1966  Rotation converged in four iterations; Cross-loading items and items loading < .35 have been discarded. Table 5 shows exploration analyses of the three different factors. It is interesting to notice that the mean response on "Methods of Black Pedagogy" is clearly lower than the one on "Values of Black Pedagogy", as expected. literature is taken into account, but it has to be considered both in the light of the context of sample recruitment and of the fact that age distribution in the pilot study was very highly skewed towards lower values.
Item 42 is dedicated to temporal collocation and presented results that were quite challenging, since 61.1% of participants clearly agreed with this collocation, but the 38.9% was not completely satisfied with the temporal definition proposed by the item. This result suggested a reformulation of the item to help identifying what period of time participants have in mind when referring to the pedagogical practices used "in the past".
The comparison between the diffusion of Black Pedagogy's practices in the past and nowadays revealed that participants report a general decrease in the use of such educational measures except for what concerns the practices of blackmailing to control children actions (method "m" in Table 6) and of justifying unpleasant educational measures by telling the child that these are applied for his/her own good (method "n" in Table 6). Table 6 summarizes the mean response for each Black Pedagogy method and the significant level of average differences in the comparison between past and present diffusion. Note. 1 = "Not Present at all"; 2 = "Present, but not Common"; 3 = "Present"; 4 = "Widespread".
Observing average comparisons, it emerges that mean differences are less pronounced for those educational practices that do not involve a physical type of harm to the child, culminating in the approximatively same level of current diffusion for what concerns method "m" and in a not significantly greater diffusion today than in the past of method "n".
Finally, bivariate correlations have been performed between the three factors and the results on the lists of ED section. The only significant correlation was found between the scores on "Values of Black Pedagogy" and the evaluation of the diffusion of Black Pedagogy practices in the past (r = .18, p < .01): a low positive correlation suggesting that those who observed a higher diffusion of Black Pedagogy practices in his/her past experience tended also to score higher on the "Values of Black Pedagogy" factor. One possible interpretation of such result could be that those who have been in contact (even only as observers) with such practices tend to assimilate Black Pedagogy's values and objectives, thus supporting the idea of an intergenerational transmission of physically and mentally violent disciplinary practices (Miller, 2007;Perticari, 2016). Since the Florio,Caso,& Castelli 339 sample of pilot study was homogeneous with regard to gender, educational qualifications and age, no specific analyses were conducted involving these variables.

Results of the Main Study
Following the results of the pilot study, the Black Pedagogy Scale has been adjusted to elaborate a final version, which was administered to the participants of the main study (N = 830). Analyses on this second dataset replicate the exploration of scale proprieties but focus in particular on the investigation of the differences between the subgroups created on the basis of demographic information. To this purpose, t-tests, bivariate correlations and ANOVAs have been performed, according to the suitability of data for each specific analysis.
The Black Pedagogy Scale in its final version (with a minimum possible score = 24 and a maximum possible score = 96 in BPO section) was administered to a sample of 830 adult subjects, parents of primary school-age children. The mean score was 60.33 (SD = 8. Pedagogy" is clearly lower than in the case of the other two factors, as emerged in the pilot study. Pearson Product-moment correlation between BPO section and age resulted negative, weak, and significant (r = -.073, p < .05), thus suggesting again that there is slight decrease in scores on BPO section with the increase of age. Nonetheless, no significant correlations have been found between age and BPO subscales.
The continuous scale of age has been collapsed into two age groups, since the cases of under 30 years and The Black Pedagogy Scale 340 over 50 years of age were very few (4.1%), and an independent samples t-test was conducted to compare scores on BPO for the age group under 40 years and for the age group over 40 years. Unequal variances have been assumed since Levene's Test for Equality of Variances was significant (p = .002), and a statistically significant difference has been found between the scores of participants aged under 40 years (M = 61.06, SD = 7.32) and those of participants aged over 40 years (M = 59.69, SD = 8.66): t(822.5) = 2.469, p = .014. Another independent-samples t-test was performed to compare BPO scores for parents of an only child and for parents of more than one child. A significant difference has been also found in the comparison of these two conditions, indicating that scores of parents of one child (M = 60.74, SD = 8.22) were significantly higher than those of parents with more children (M = 59.40, SD = 7.73): t(828) = 2.2, p = .028. A first attempt to interpret this unexpected result could be that parents of one child rely more on culturally learned methods of parenting practice, whereas the increased experience, effort, and/or relational complexity resulting from having two or more children lead parents to deviate from their educational legacy and to work on their own parenting solutions. This is a possible interpretation that surely needs to be further investigated in future studies. In fact, according to Edwards (2014) parenting experience (i.e., raising singletons or more than one child) is a variable that deserves more attention: in the context of her study, the author found that it is more likely for mothers (37.8%) or of their grandparents (36.3%) as the last one in which "old fashioned" educational practices were applied. A small proportion indicated that such methods were used until their great-grandparents generation and no further (4.2%). An unexpected 20.1% declared that respondents' own generation has witnessed the application of educational methods commonly referred to in our territory as those "old fashioned" and defined in the reference literature as Black Pedagogy practices.
Paired sample t-tests were conducted on responses concerning past and present diffusion of each Black Pedagogy method. All the differences between the mean responses resulted significant (p < .001), indicating a statistically significant decrease in the diffusion of such practices according to participants' estimations, consistently with the clear low agreement on "Methods of Black Pedagogy" subscale belonging to prior section.  Note. 1 = "Not Present at all"; 2 = "Present, but not Common"; 3 = "Present"; 4 = "Widespread". CI = confidence interval; LL = lower limit; However, if the specific amounts of such decreases are emphasized, it appears that they are not always similar for all methods. In other words, some clearly decreased more than others. In fact, it can be observed that "Pedagogical beating" is the practice that decreased the most, whereas "Blackmailing" decreased the less. In general, one can see that most methods involving a physical level decreased more than those concerning the psychological area (e.g., treating coldly, humiliating, lying, blackmailing, etc.). Unexpectedly, according to respondents, physical violence decreased less than pedagogical beating, and "Cautionary tales" is one of the methods that decreased more, contrary to what was found in the pilot study where respondents were students with a mean age of 20. More specifically, "Pedagogical Beating" is the method that theoretically could be considered the most representative of the Black Pedagogy construct. On this basis, it is to be presumed that if "Pedagogical Beating" was taken as a reference point, more harmful methods on the physical level (e.g., denial of food, toughening up, and physical violence) should be estimated as less diffused nowadays, whereas psychologically detrimental methods should appear more widespread. As can be seen in Table 9, this supposition The Black Pedagogy Scale 342 is confirmed with the exception of the "Humiliating" method, which seems diffused as much as "Pedagogical Beating". All the other estimations of diffusion are statistically different from the one regarding "Pedagogical Beating": according to respondents, physically harmful methods are less diffused (positive mean difference), whereas psychologically detrimental disciplinary practices are more diffused (negative mean difference) than "Pedagogical Beating". Finally, bivariate correlations have been conducted between BPO section total score, its subscales, and the results on the doubled 12-items list of educational practices used in the past and at the present day. Results are summarized in Table 10. As can be seen, correlations do not resemble the results of pilot study, probably because university students do not have children yet. In fact, when parents are responding, a weak negative correlation emerges between the estimation of Black Pedagogy diffusion in the past and the agreement with its methods. It appears that those parents who in the past witnessed a greater diffusion of the educational practices typical of Black Pedagogy, Florio,Caso,& Castelli 343 tend to show less agreement with Black Pedagogy methods at the present day and, presumably, with their application to their own children. A result that is in line with the fact, already presented above, that scores on BPO seem to become lower with the increase of age and of the number of children.

Discussion and Conclusions
The Furthermore, even if our results suggest that physically maltreating educational practices of Black Pedagogy are not accepted nowadays, the same cannot be said for the educational values and objectives from which such practices consistently originated. The ongoing diffusion of the subtlest forms of disciplinary methods that are definable as psychologically harmful could be therefore due to the persistence of an obsolete hierarchical and authoritarian conception of the right way to raise and educate children. Nowadays, the threshold of "acceptable" mental or physical violence in an educational relationship seems to be lower than in the past, but the impression is that not all forms of violence are subjected to the same rate of decrease in their use. In fact, responses on the doubled 12-items list regarding estimations of diffusion made by participants, give the impression that psychologically harmful educative and disciplinary practices have not decreased as much as physical ones. In our opinion, this result suggests two possible scenarios: if adults are pursuing educational ideals without being aware that these are consistent with abusive disciplinary methods, they could either apply a level of disciplinary violence deemed acceptable in their cultural and social context (e.g., a slap, a verbal insult, etc.) or, in the best case, they could feel deprived of the means to carry out their educational duty towards children. Both these possibilities indicate that if a change at the level of child-rearing and educational practices is desired, intervention should be placed at the level of values and objectives in order to change them and to allow the spontaneous emergence of different methods. Such consideration could have important implications in interventions aimed at promoting healthy relationships between adult and children. Moreover, The Black Pedagogy Scale 344 the fact that a higher educational level is associated with lower scores on BPO section could help in identifying populations that are most at risk and in designing targeted intervention aimed at interrupting the intergenerational transmission of such practices and values.
The present study shows some limitations that suggest some possible lines of future research. In fact, concurrent validity has still to be assessed, as well as the reliability of the instrument with subjects of different cultures.
As mentioned above, the topic concerning the difference in scores on the BPO section between parents raising an only child and parents of more than one child deserves a further in-depth investigation. Moreover, it could be worth exploring the responses of parents with children of different age groups (e.g., preschoolers, adolescents, etc.), and of adults who work in educational contexts with different roles (e.g., teachers, educators, sport trainers, child advocates, etc.) in order to use Black Pedagogy as a further interpretative construct of educational practices, thus designing possible projects of intervention in different contexts.

Funding
The authors have no funding to report.