PHYSICAL ABUSE: BIAS, EXAGGERATION AND RIGGED STATISTICS
In this part of our paper, we analyse the claims in the Report (http://wcd.nic.in/childabuse.pdf) about physical abuse of children in the family. The Report also covers physical abuse of children in schools and abuse of working children, street children and institutionalised children. But we focus on the claims about abuse by family members, as the most insistent claim of the Report is that Indian families are inherently violent towards children.
MISAPPLICATION OF STATISTICAL PRINCIPLES AND POOR QUALITY OF DATA
Since the data underlying the Report was obtained through purposive and other non-probability sampling, for the reasons discussed in the earlier part, it was wrong to apply its findings to the general population of children in India. So, for example, the Report is not justified in making the claim that two out of every three children in India are “physically abused” based on its finding that about 69% of the child respondents in the survey were reported to have been “physically abused”.
This restatement of a finding of 69% abuse as two in every three children being abused is in of itself a statistical misrepresentation. Such a statement purports to express the manner in which a phenomenon is distributed in a population. But the percentage rate of a phenomenon is not the same as its distribution. Certainly, you cannot assume that a phenomenon is uniformly distributed based on its rate of occurrence. If random sampling is used, then you may be able to make such a claim, always subject to stated limitations and margins of error. But, as already discussed, random sampling was not used in the data gathering for the survey.
To put it very simply, if you are looking at 10 families with, say, two children each, a finding that 50% of the children love chocolate does not mean that one child of every two you look at in the sample will turn out to love chocolate. If you randomly pick children in twos from the sample, some two may both Report disliking chocolate, and some two may both Report loving it. It is important to bear this point about distribution in mind, because implying that a horrible act like child abuse is uniformly found in a population compounds the accusation. It implies a situation where practically every family has at least one abused child. It implies, not just the existence of child abuse, which no one would deny, but of a nation of abusers.
It might be said that with a high rate of 69% physical abuse being reported, the fact that the incidence of abuse may not be uniformly distributed is not of much significance. But, as it emerges below, this rate itself, along with all the other percentages and ratios claimed in the Report, are highly suspect.
MISREPRESENTATION AND EXAGGERATION OF OWN DATA
The data gathering for the Report was unsystematic, arbitrary and unscientific. The reasons for this are explained in detail in the earlier part, and are not repeated here. But the problem in the Report is not just the failure to apply the proper statistical procedures. Even ignoring the Report’s failure to gather credible data, the Report misrepresents its own data and exaggerates its findings in a number of ways.
OVERBROAD DEFINITION OF “PHYSICAL ABUSE”
The first is by using alarmist language such as “abuse”. “Child abuse” in the common understanding refers to a situation where the perpetrator is an adult personally known and trusted by the child. Hence the term “abuse”, rather than a more general term such as “violence” or “assault”.
To say that the overwhelming majority of children in India are “physically abused” conveys an impression about what happens to Indian children, not in the public sphere or in their interaction with playmates, strangers or police officers, but in their personal, everyday life, in their homes, schools and communities. It conveys an impression of aggression and violence perpetrated, not by playmates or siblings, but by the child’s adult custodians. Above all, since “abuse” is generally understood as referring to things happening in a child’s family, it is a direct accusation levelled at Indian parents.
So the first manipulation in the Report is to use the expression “physical abuse” for instances where the perpetrator is not an adult known to the child, but another child, such as a playmate or sibling; or a stranger, such as a policeman.
Bullying by peers and police atrocities are reasonably a concern for anyone looking at the well-being of children. But, given the general understanding of the term “child abuse” as applying to what is done to a child by its adult intimates, it is blatantly misleading to club peer bullying and police atrocities with parental or school abuse. The last thing on the mind of anyone reading a Report saying that two out of every three Indian children are physically abused, is that the perpetrator of the abuse was a class fellow, stranger or police officer.
The exaggeration becomes even more apparent when you realise that the so-called “physical abuse” of the Report includes, not just severe beating or causing injury, but “pushing”, “slaps” and “beating” with a “stave” or “stick”.
Some people hold that children should never be slapped or spanked, but even allowing for that view, there is no warrant for labelling slapping as “abuse”.
Serious allegations that the Report makes, such as that two out of every three children, or 69% children are physically abused in India, come across very differently when you realise that by “physically abused” the Report means not just severe beatings, but a whole range of acts from slapping, to disciplining by corporal punishment, to injury-causing assaults, in which (as we will observe later) the largest proportion of so-called abuse is slapping.
DATA MISREPRESENTED TO PORTRAY INDIAN PARENTS AS HIGHLY PHYSICALLY ABUSIVE
The Report paints a picture of deep oppression, bordering on hatred, of Indian families, especially parents, towards their children. We are told that “[Indian] fathers and mothers, consider their children as their property and assume a freedom to treat them as they like”; that “[Indian parents] adopt harsh methods of disciplining children”; that “children faced high level of physical abuse in families”; and that “often the child is the easiest target for the parent to vent their frustration on”.
Let us take a look at the evidence for these allegations.
REPORTED RATE OF 89% PARENTAL ABUSE
The Report states that “89% children were subjected to physical abuse by parents”. Recall that the overall rate of physical abuse of children by family and others combined was stated in the Report to be 69%. So the claimed rate of parental abuse was clearly not calculated on the total number of children surveyed, but on a smaller sub-set of them. Stating the rate by looking at a smaller subset has the obvious effect of magnifying the figures. So it is worthwhile to investigate how the 89% figure was arrived at in the Report.
With its characteristic lack of analytical rigor (or was it deliberate obfuscation?), the Report does not state how it arrived at its claimed rate of 89% parental abuse. But other figures in the Report give us some hints.
The Report says: “An attempt has been made to see the extent of physical abuse of children in families as compared to the physical abuse of children by others. The study revealed that the percentage of physical abuse inflicted by family members (48.7%) was higher than that of others (34.0%).” Again, it is not stated as to how this percentage was calculated. But since “parents” are a sub-set of “family members”, the overall rate of parental abuse cannot be higher than the overall rate of family member abuse. How then does the Report claim a figure of 89% parental abuse, while at the same time claiming nearly half that rate, 48.7%, for family member abuse?
We will assume that the figure of 89% abuse by parents was calculated on the number of children reporting abuse by a family member. In other words, the number of children reporting parental abuse was 89% of (48.7% of X), i.e., 43.34% of X, where X could be the total number of children surveyed, or some smaller sub-set of the total (such as the total number of children in the survey that had family (i.e. were not orphans).
So what appears to be a staggering claim that “89% children were subject to physical abuse by parents” turns out, if our assumption is correct, to be supported by data that only evidences below half of that figure, 43.34% children (whether overall, or a smaller sub-set of those surveyed), reporting physical abuse by parents.
Even though 43.34% is less than half the rate claimed by the Report, it would appear to indicate a fairly high percentage of parental physical abuse, until we recall the overbroad definition of physical abuse used in the Report. Since “physical abuse” is broadly defined in the Report to include non-severe acts, such as pushing and non-severe slapping, we are not looking at data that shows 43.34% parents severely beating or causing injury to their children. We might even be looking, based on the discussion below, at data that merely tells us of a substantial amount of non-severe slapping by mothers of children.
DATA REVEALS SEVERE CASES OF ABUSE ARE A MINORITY
For some reason, the Report does not give the overall breakdown of forms of physical abuse by parents. After setting out the overall figures discussed above, the Report goes on to give familial physical abuse rates only for a sub-set of the child respondents, a group of 2245 children classified as “Children in family environment, not going to school”.
We will not apply the rates of abuse found in the sub-set to all the child respondents claiming family abuse, as children in other sub-sets (termed “evidence groups” in the Report) of the sample surveyed may also have claimed family abuse. But if the data on all child respondents claiming family abuse followed the same general pattern as the data on family abuse revealed in the sub-set, then we are looking at a very different situation as to family child abuse to the one portrayed by the Report.
Let us first look at how much severe physical abuse is reported in the sub-set as having been caused by parents. If most of the parental abuse reported is not severe, then we are looking at highly unjustified claims by the Report about Indian parents.
In the sub-set, 14.83% of the reported in-family abuse is said to be severe. So of the sub-set used to study familial abuse, severe abuse occurred only in a minority of cases. We will assume that the relatively smaller incidence of severe abuse by family members (and thus by parents) within the sub-set also applied when you looked at all the children reporting family abuse in the survey.
Data reveals the majority of cases reported as abuse are cases of slapping.
The Report on the sub-set goes on to say that 74.3% of in-family abuse was in the form of “slapping/kicking”.
Another way the Report misleads the reader is by clubbing the data on slapping and kicking of children. Slapping and kicking have been treated as one category of physical abuse even though the two are widely different in severity and impact on the child.
In the questionnaire for children being asked about physical abuse by family members, slapping and kicking are presented as one category. So a child answering “yes” for slapping, is also reported as answering “yes” for kicking, which is highly misleading. In the questions for children on physical abuse by “others”, slapping and kicking are presented as two distinct categories, but the Report does not disclose the numbers reporting kicking versus slapping.
It is not clear why slapping and kicking were clubbed except to make findings about slapping, which is a relatively minor category of “abuse”, look like a finding about kicking, an ugly and more severe form.
We have some clues as to the relative proportion of slapping to kicking from the data on young adults surveyed about their childhood experiences. Forty-nine percent young adults are reported as having been beaten in childhood, the breakdown of forms of so-called “physical abuse” being slapping at 44%; kicking at 7%; beaten by stick at 36%; and pushing at 8%.
So what we have here is data indicating that the incidence of slapping was overwhelmingly higher, to the point of there being no comparison, to the incidence of kicking.
If we assume that this same relative disproportion applies to child respondents reporting slapping/kicking by family members, then what we are looking at is data that evidences that among the children reporting being beaten by family members, the overwhelming majority being beaten in the form of non-severe slapping. Since parents are a sub-set of “family members”, we will assume that the same logic applies to them.
OVERALL PICTURE ON PHYSICAL ABUSE FROM REPORT’S DATA
Let us now step back and look at the overall picture that emerges:
• parental physical abuse rates were half or less than half of what was claimed in the Report
• of the cases categorised as parental physical abuse, the majority were cases of light slapping
• severe family-member, and therefore parental, physical abuse is reported only in a minority of cases
• 31% children interviewed reported no physical abuse, not even light slapping and caning.
WE DO NOT GIVE THE ABOVE ANALYSIS TO MAKE ANY CLAIMS AS TO THE REAL RATES OF PARENTAL ABUSE IN THE COUNTRY BECAUSE, AS EXPLAINED ABOVE, WE DO NOT BELIEVE THE DATA IS RELIABLE. WE ARE HERE MERELY ANALYSING THE DATA TO DEMONSTRATE HOW THE REPORT EXAGGERATES FIGURES AND DOES NOT SUPPORT THE CLAIMS OF THE REPORT ABOUT WIDESPREAD PHYSICAL ABUSE OF CHILDREN IN INDIAN FAMILIES.
DIFFERENCE BETWEEN DISCIPLINING AND ABUSING A CHILD
In the questionnaire used for child respondents, the entire issue of physical abuse is set up as a question of how adults are disciplining their children. So we can reasonably assume that in almost each reported case of slapping, the child was not talking about wanton and malicious slapping, but of being slapped in the context of being corrected or restrained from doing something that was bona fide believed by the adult in question to be wrong or harmful.
Again, some people hold the view that any kind of slapping of a child, whether to discipline it or otherwise, is wrong. But even allowing for that point of view, when the data showed the majority of children do not Report parental physical abuse, and that most cases of physical abuse were cases of a parent slapping with the intent of correcting an errant child, the harsh condemnation of parents in the Report is totally unjustified.
STATISTICS ON PARENTAL BEHAVIOUR TO BE UNDERSTOOD IN CONTEXT OF THEIR BEING THE PRIMARY ACTORS IN THE LIFE OF THEIR CHILDREN, ESPECIALLY WHEN CHILDREN ARE YOUNG
Another point that must always be borne in mind when assessing data regarding the experiences of children is that in any survey of childhood experiences, no matter how you design it, the proportion of parents reported as so-called perpetrators is likely to be higher than any other class of persons.
This would be seen especially in cases where the survey is of young children (recall that the biggest group of children surveyed for the Report was of the age group five to 12, so the survey included more younger than older children). The reason is simple, unless you are looking at orphans, most children live with their parents and the overwhelming proportion of children’s direct experiences are either with parents, or mediated through parents.
So whether it is kissing or slapping, praising or scolding, encouraging or restricting, the maximum instances of such experiences will often be, as a proportion, higher from parents than from other categories of persons. The younger the age of the child, and especially if you are going to restrict your research to “disciplining” of children, the higher the proportion of parental actors you will find.
Since parents are the prime actors in a child’s life, to say that “mostly it is parents who slap their children” is, statistically speaking, no more significant than saying that “mostly it is parents who take care of their children” and it is not the same as saying that “most parents slap their children”.
Suranya Aiyar was a practising lawyer before opting to become a stay at home mother in 2010. Since 2012, she has been writing and critiquing Western-style child protection laws as advocated in India by UNICEF and Save the Children. On a pro bono basis she has given support and advice to Indian families facing confiscation of their children abroad by child protection authorities. She also writes and illustrates children’s books.
To be continued