Age Heaping Research Articles

Domesticating Human Capital

This article investigates the consumption of knowledge in early modern Amsterdam. A dataset of seventeenth- and eighteenth-century probate inventories is employed to examine the synergies and inequalities between the skilled and educated middle classes and the intellectual elites. A series of democratisation waves in the ownership of books, writing equipment, and measuring tools confirms the unprecedented levels of basic literacy and numeracy skills in the urban centres of the early modern Low Countries revealed by research on signature proficiency and age heaping. The concentration of secular books and advanced knowledge objects in the hands of a small but growing group of affluent households, on the other hand, corresponds to other research that has fixated instead on the role of upper-tail human capital in scientific, technological, and economic progress. Yet, the relatively low value estimates of libraries and scientific instruments, together with a more qualitative examination of two amateur scientists of middling background, dovetails with the hypothesis that the Dutch Scientific Revolution and Enlightenment were marked by a close interaction and mobility between craftsmen and scientists. An above-average income and enough leisure time to develop intellectual interests could be sufficient for inhabitants of Amsterdam to cross the Rubicon from consuming to (re)producing knowledge.

The expansion of basic education during ‘deskilling’ technological change in England and Wales, c. 1780–1830

AbstractThe first country to industrialize – England – ostensibly did so without expanding investment in the basic education of its workforce. The empirical evidence underpinning this argument for England rests largely on signature rates at marriage. These are not a perfect indication of educational achievement, particularly as many children never learned to write. More problematically, I argue signatures are likely to have systematically underestimated human capital in industrial districts. In place of signature data, I propose age heaping, a measure widely understood as a proxy for numeracy but shown here to be closely related to both reading and writing abilities. In contrast to signatures, this measure suggests that ‘deskilling’ industrialization induced human capital accumulation. I argue that this occurred not because human capital was directly productive, but rather because schools provided a valuable signal. Sunday school attendance signalled low leisure‐preference among child workers and were popularly attended in industrial districts. Further, such schools taught children to read but not write, which they considered inappropriate for the Sabbath, accounting for the discrepancy between these two measures of human capital.

Revisiting the Recommended Duration of Interviews Conducted by Mobile Phone in Low- and Middle-income Countries: A Randomized Trial in Malawi

Guidelines for conducting surveys by mobile phone calls in low- and middle-income countries suggest keeping interviews short (<20 minutes). The evidence supporting this recommendation is scant, even though limiting interview duration might reduce the amount of data generated by such surveys. We recruited nearly 2,500 mobile phone users in Malawi and randomly allocated them to 10-, 20-, or 30-minute phone interviews, all ending with questions on parental survival. Cooperation was high in all groups, and differences in completion rates were minimal. The extent of item nonresponse, age heaping, and temporal displacement of deaths in data on parental survival generally did not vary between study groups, but reports of maternal age at death were more reliable in longer interviews. Recommendations about the duration of mobile phone interviews might be too restrictive. They should not preclude additional modules, including ones on mortality, in mobile phone surveys conducted in LMICs.

Age reporting by and for older people in Uganda: relationships with frailty, human capital and population registration

In this article, we investigate the accuracy of age reporting by people aged 60 and older and proxy reporting by their carers in a peri-urban area of Uganda, and analyse the factors that influence reporting by both groups. We find a high level of age heaping on terminal digits 0 and 5, indicating poor knowledge of age. Contrary to other studies, we find that literate people were more likely to exhibit age heaping. We link this to the absence of birth registration for this cohort and the introduction of National Identification (ID) cards in Uganda five years before our survey. We conclude that age heaping is better interpreted as an indicator of registration machinery than of human capital. We also find that the health, functional capabilities and education of older people influenced the age ascribed to them by their carers. Carers who knew the older people less well were more likely to overestimate their age, and carers of healthy and more educated people were more likely to report a younger age than that reported by the older people themselves. Where people don’t know their age, the age they report may also be influenced by their health and capabilities, making it difficult to establish true relationships between chronological age and outcomes such as health. In many disciplines, self-reported age or age reported by proxy respondents is accepted uncritically by researchers, but our study shows that in peri-urban Uganda age reporting remains approximate and biased, and this has strong implications for appropriate targeting and monitoring of interventions to support healthy ageing in such contexts.

Comparison of Kernel Smoothing and Local Polynomial Smoothing Method in Overcoming Age Heaping

Age data plays an important role in every aspect yet there are found age misreporting. It involves digit preference that causes build up in a certain age. Digit preference in demography is called age heaping that often happens at age with 0 and 5 as the last digit. Age heaping induces poor data quality and data bias that could influence government policy making. Two indicators used to detect age heaping are Whipple Index (WI) and Myers Blended Index (MBI). Methods to cope with age heaping are nonparametric regression approaches which are Kernel Smoothing and Local Polynomial Smoothing. The objective of this research is to measure and elevate the quality of population age data and population mortality data in Sensus Penduduk (SP) 2020 as well as comparing methods between Kernel Smoothing and Local Polynomial Smoothing. The data being used in this paper is SP2020 which the research variables are age population, age of death, and total population. The result shows that the data quality of total population death is inaccurate compared to total population thus needs a smoothing process to improve age data to population data accuration. The method that has better accuracy is the Local Polynomial Smoothing method.

Numeracy and consistency in age declarations: a case study on nineteenth and twentieth century Catalonia

A rich tradition exists on measuring human capital in historical societies through basic mathematical skills using the method of age heaping. In this article, we novelty propose to calculate it also through the analysis of the consistency in individual consecutive age declaration statements in five towns of Baix Llobregat’s county (Barcelona, Spain) during the nineteenth and early twentieth centuries. The region underwent a proto-industrialization in the eighteenth century and was one of the main wool and cotton textile producers in the nineteenth century in Catalonia. Our main results show that (I) numeracy levels in this county were above the Spanish average in the whole period and, during the first years of the nineteenth century, ahead of countries such as Italy, Portugal and the USA. However, its numeracy dropped from 1880’s due the early industrialization of Catalonia in the Spanish and Southern Europe context, the existence of child labor in the factories and that also coincided with the phylloxera crisis. (II) To this effect, women presented lower levels of numeracy than men, but their levels tended to converge. (III) As regards emigrants, those who came to the rest of Spain had the lowest level of numeracy compared to the natives. (IV) Moreover, we found that literacy, the 1901 education reform and a higher socioeconomic status had a positive effect on reporting ages consistently over time. (V) Finally, we prove a positive association between systematically reporting ages ending in 0 or 5 and having a higher individual error in age statements along the life course.

Evaluation of Age-Sex Data Collected in the 1991, 2001 and 2011 Population and Housing Censuses of Namibia

This paper examined the accuracy of age-sex data reported in the Namibia population censuses with the aim to establish patterns and trends in the quality of age-sex reporting. To date, Namibia has conducted three population and housing censuses. However, it is not clear if there has been any advancement in the quality of data collected on age and sex over the years. The study followed a cross-sectional study design using the 1991, 2001 and 2011 population censuses data from Namibia Statistics Agency. We computed Whipple’s and Myers’s Blended indexes to measure age heaping. The age-sex ratio and UN joint score were used to ascertain the possibility of population undercounts and displacement between age groups. The results show that age heaping was fairly accurate in all three censuses. Analysis of single-age distribution showed a minimal digit preference and avoidance across all three censuses. Likewise, Myers blended index portrayed that terminal digits 0, 1, 2, 5, 8 and 9 were preferred the most by both sexes in all three censuses. The Whipple’s Index showed a two-point decline from 106 in 1991 to 104 in 2001 and 2011 suggesting an improvement in age reporting. The UN joint score for Namibia found that the data collected in all three censuses were inaccurate in terms of age-sex displacement.

Increased Age Heaping in Mobile Phone Surveys Conducted in Low-Income and Middle-Income Countries

Since the beginning of the coronavirus disease 2019 pandemic, the number of surveys conducted remotely by mobile phone in low-income and middle-income countries has increased rapidly. This shift has helped sustain data collection despite restrictions on mobility and interactions. It might also allow collecting data more frequently on important demographic and socioeconomic topics. However, conducting interviews by mobile phone might affect the accuracy of reported data, for example, if respondents have difficulties understanding questions asked remotely, or data collectors have less time to probe and cross-check answers. In this visualization, the authors explore time trends in age heaping, a strong signal of reporting errors, in six African countries. They show that mobile phone surveys have generated noisier data on age than recent household surveys and censuses, thus possibly affecting researchers’ understanding of demographic processes and confounding multivariate analyses of socioeconomic outcomes.

Effectiveness of modular approach in ensuring data quality in large-scale surveys: Evidence from National Family Health Survey – 4 (2015–2016)

This study aims to examine the effect of administration of shorter and longer versions of questionnaires on key indicators such as age displacement, birth displacement, age heaping, and skipping questions on antenatal care (ANC) visits and use of contraceptive methods in India using National Family Health Survey (NFHS)-4 data. At the individual level, the effect of the adoption of the shorter and longer versions of the questionnaires on the age displacement of women and children and skipping of the key questions is insignificant. However, the results from the two-level logistic regression model reveal that at the primary sampling unit (PSU) level, work pressure, depending on the number of eligible women in a household, emerges as a confounder in skipping certain questions, namely ANC [1.18 (p < 0.09)] and contraceptive use [AOR = 1.17 (p < 0.05)]. To expand the coverage of NFHS in providing state- and district-level estimates since 2015, the overall sample size was increased from 88,562 households and 89,777 eligible women in 1992–93 to 6,01,509 households and 6,99,686 eligible women in 2015–16. As a strategy to reduce workload and non-sampling errors during the survey, a nested design and modular approach were adopted to provide estimates of maternal and child health indicators at the district/state level and sexual behaviour, HIV/AIDS, and women's empowerment at the state level. It was hypothesised that a longer version of the questionnaire canvassed in the state module may be detrimental to data quality issues. The findings of this study establish the effectiveness of adopting a modular approach in large-scale surveys, depending on the scale of investigation. However, the differential workload calls for expanding the duration of surveys in PSUs, where the number of eligible women is higher. State level variation in the key data quality indicators may be partially explained by differentials in the training of investigators by the agency and use of translators.

Age heaping and its discontents: A response to Baten, Benati, and Ferber

AbstractThis note is a rejoinder to Baten, Benati, and Ferber. We reiterate that, on close inspection, nineteenth‐century Italian census data contain a number of anomalies that sit uncomfortably with a straightforward age‐heaping‐as‐numeracy interpretation. In particular, we respond to Baten, Benati, and Ferber on the technical matters they have raised; then we show that our findings are robust to such criticism. Finally, we conclude with some general reflections on age heaping as a numeracy indicator.

Age Heaping and Life Course Reckoning

Human capital, and numerical abilities in particular, are a crucial resource for economic development, but measuring numeracy is a major challenge. The Spanish Inquisition offers a possibility to assess numerical skills, as the inquisitors performed indirectly a mathematics test with the accused. They asked for their age and for a detailed life narrative, with many defendants failing to notice if the sum of the narrative did not add up to the given age. This also allows for exploring correlations with age-heaping-based numeracy estimates at the individual level. Those who reported rounded ages tended to perform worse in “life course reckoning” tasks. Persons whose life reckoning was off by more than five years showed a 23 to 26 percent higher age heaping (and hence, lower numeracy).

Estimating the infant mortality rate from DHS birth histories in the presence of age heaping.

BackgroundThe infant mortality rate (IMR) is a critical indicator of population health, but its measurement is subject to response bias in countries without complete vital registration systems who rely instead on birth histories collected via sample surveys. One of the most salient bias is the fact that child deaths in these birth histories tend to be reported with a large amount of heaping at age 12 months. Because of this issue, analysts and international agencies do not directly use IMR estimates based on surveys such as Demographic and Health Surveys (DHS); they rely instead on mortality models such as model life tables. The use of model life tables in this context, however, is arbitrary, and the extent to which this approach appropriately addresses bias in DHS-based IMR estimates remains unclear. This hinders our ability to monitor IMR levels and trends in low-and middle-income countries. The objective of this study is to evaluate age heaping bias in DHS-based IMR estimates and propose an improved method for adjusting this bias.Methods and findingsOur method relies on a recently-developed log-quadratic model that can predict age-specific mortality by detailed age between 0 and 5. The model’s coefficients were derived from a newly constituted database, the Under-5 Mortality Database (U5MD), that represents the mortality experience of countries with high-quality vital registration data. We applied this model to 204 DHS surveys, and compared unadjusted IMR values to IMR values adjusted with the log-quadratic model as well as with the classic model life table approach. Results show that contrary to existing knowledge, age heaping at age 12 months rarely generates a large amount of bias in IMR estimates. In most cases, the unadjusted IMR values were not deviating by more than +/- 5% from the adjusted values. The model life table approach, by contrast, introduced an unwarranted, downward bias in adjusted IMR values. We also found that two regions, Sub-Saharan Africa and South Asia, present age patterns of under-5 mortality that strongly depart from the experience represented in the U5MD. For these countries, neither the existing model life tables nor the log-quadratic model can produce empirically-supported IMR adjustments.ConclusionsAge heaping at age 12 months produces a smaller amount of bias in DHS-based IMR estimates than previously thought. If a large amount of age heaping is present in a survey, the log-quadratic model allows users to evaluate, and whenever necessary, adjust IMR estimates in a way that is more informed by the local mortality pattern than existing approaches. Future research should be devoted to understanding why Sub-Saharan African and South Asian countries have such distinct age patterns of under-five mortality.

Age heaping pattern and data quality: evidence from Indian Household Survey Data (1991–2016)

Age structure is crucial to various socio-economic and demographic factors. Better age quality data are critical to determine the reliability of large scale household surveys and the estimates they provide. Considering the importance and large sample these household surveys represent, we examined one of India’s largest household survey (NFHS) to determine its data quality and accuracy. Using multiple indices we also made a comparison of various age reporting measures to assess pattern of age heaping. Our results found that age heaping is a gray concern for household surveys in India. Though there has been some improvement, but over all age quality data are still very rough, affecting likely the survey estimates. Females are performing better on contrary to what earlier studies have found, which is likely due to improving socio-cultural and living norms in Indian settings. It is clear that age preference can likely cause the substantial biases in demographic factors and survey estimates. Thus studies examining age structure or any such methods where age is significantly important should adjust the age bias before carrying out further analysis. Furthermore, a single approach for age adjustment techniques is must to improve the age quality data for better policy implications.

Rethinking age heaping: a cautionary tale from nineteenth‐century Italy†

AbstractAge heaping is widely employed as an indicator of human capital, more specifically of numeracy. We re‐examine the age heaping–numeracy link in the light of evidence from nineteenth‐century Italian censuses, in which education explains little of the variation in age heaping. We argue that in general age heaping is most plausibly interpreted as an indicator of cultural, economic, and institutional modernization rather than a straightforward measure of individual cognitive skills. We do not rule out the use of age heaping as a numeracy indicator, but this needs to be done with research designs that are alert to historical specificities of the context under investigation.

Evaluating the Demographic and Health Surveys Mode Switch From PAPI to CAPI: An Experiment From Turkey

The Demographic and Health Surveys (DHS) have been carried out in over 90 countries since 1984, as interviewer administered household surveys conducted initially by paper and pencil interviews (PAPI). Computer assisted personal interviews (CAPI) were introduced in the 2004 Peru DHS, and since then numerous countries have also switched. However, DHS randomized mode comparisons have been limited. The 2018 Sixth Turkey DHS was conducted using CAPI but allocated one household from 21 in each of 754 clusters to PAPI. This analysis examines a wide range of potential differences between modes: interviewer attitudes toward modes; response rates, underreporting and misreporting of persons or events, number of selections to “check all that applies” questions, respondents' attitudes towards modes reflected by responses to sensitive questions, satisficing behavior such as age heaping, straight-line response patterns, and use of don’t know options; and some operational aspects of modes such as retrospective monthly contraceptive prevalence rates, presence of others during interview, and interview length. Findings show that, despite strong interviewer CAPI preference, CAPI and PAPI were on average almost identical in terms of responses. CAPI took 11 min less (total duration of 33 min). Analysis of retrospective monthly contraception use indicated potential underreporting by CAPI for past use, an issue highlighted before in DHS literature. Overall, the switch to computer technology in DHS surveys does not appear to change estimates or levels of nonsampling errors, although some differences with respect to PAPI mode may need DHS designer attention.

Quality of age data in the Sierra Leone Ebola database.

IntroductionWhile it is suspected that some ages were misreported during the 2014-2016 West African Ebola outbreak, an analysis examining age data quality has not been conducted. The study objective was to examine age heaping and terminal digit preference as indicators for quality of age data collected in the Sierra Leone Ebola Database (SLED).MethodsAge data quality for adult patients was analyzed within SLED for the Viral Hemorrhagic Fever (VHF) database and the laboratory testing dataset by calculating Whipple´s index and Myers´s blended index, stratified by sex and region.ResultsAge data quality was low in both the VHF database (Whipple´s index for the 5-year range, 229.2) and the laboratory testing dataset (Whipple´s index for the 5-year range, 236.4). Age was reported more accurately in the Western Area and least accurately in the Eastern Province. Age data for females were less accurate than for males.ConclusionAge data quality was low in adult patients during the 2014-2016 Ebola outbreak in Sierra Leone, which may reduce its use as an identifying or stratifying variable. These findings inform future analyses using this database and describe a phenomenon that has relevance in data collection methods and analyses for future outbreaks in developing countries.

Gender Gap in Numeracy and the Role of Marital Status. Evidence from 19th century France

Les indicateurs de capital humain conventionnellement utilisés dans les études de long terme ( e.g. taux d’alphabétisation, niveau d’instruction, nombre d’années de scolarisation) sont rarement disponibles par statut marital, voire inexistants. Un indicateur de capital humain fréquemment utilisé pour les périodes antérieures au 20ème siècle est l’« attraction des âges ». Nous utilisons cette méthode pour analyser et représenter graphiquement l’hétérogénéité des capacités cognitives et différences observées en termes de genre et par statut marital en France au milieu du 19 ème siècle. Les indices calculés montrent que les femmes mariées « accumulent » significativement moins que les femmes célibataires, alors que les hommes mariés « accumulent » davantage que les hommes célibataires. Nous explorons la validité de quatre hypothèses pour expliquer ces différences. Nos résultats montrent que les données concernant les femmes mariées et célibataires sont sujettes à des biais et qu’elles doivent, en conséquence, être utilisées avec précaution pour mesurer les compétences de bases en numératie. Nous trouvons malgré tout qu’il existe une corrélation robuste entre taux d’alphabétisation des individus et capacités en numératie. Nos résultats suggèrent que les aptitudes en calcul des femmes mariées sont davantage corrélées avec les indicateurs classiques de capital humain, comme le taux de scolarisation, que ne le sont les aptitudes des femmes célibataires et celles des hommes.

Quantifying the Potential Interviewers’ Effect in the National Family Health Surveys in India

There is little research on quantifying the potential interviewers’ effect in low- and middle- income countries including India. This study examines interviewers’ effect in two large-scale National Family Health Surveys in India in 2005-06 and 2015-16, with a particular focus on whether the interviewers’ effect varies by the sensitivity level of the questions that are asked in the surveys. We conclude that the interviewers’ effect varies by the sensitivity level of the outcome, with a large interviewers’ effect on responses to potentially sensitive questions (such as the woman’s justification of wife beating in specific circumstances), but only a very small or negligible effect on responses to non-sensitive questions (educational attainment, age heaping, and modern contraceptive use) in both the surveys. The results were similar using two alternative analytical approaches (deviance residuals and a cross-classified random intercept logit model). The interviewers’ effect was much larger in the 2015-16 survey, which had a sample size of 699,686 women age 15-49, than it was in the 2005-06 survey, which had a sample size of 124,385 women age 15-49.

NUMERACY OF RELIGIOUS MINORITIES IN SPAIN AND PORTUGAL DURING THE INQUISITION ERA

ABSTRACTWe assess the numeracy (age heaping) of religious minorities, particularly Jews, and other defendants of the Spanish and Portuguese Inquisitions, and compare it with the general Iberian population. Our database includes 13,000 individuals who took part in Inquisition trials, and 17,000 individuals recorded in censuses and parish registers who serve as a control group. We thoroughly discuss the representativeness of our samples for the populations we aim to capture. Our results point at a substantial numeracy advantage of the Judaism-accused over the Catholic majority. Furthermore, Catholic priests and other groups of the religious elite who were occasional targets of the Inquisition had a similarly high level of numeracy.

Микропереписи населения России: прошлое, настоящее и будущее

Прослежена история трех микропереписей населения, проведенных в России в 1985, 1994 и 2015 годах. Выделены их организационно-методологические характеристики, программные особенности. Поскольку микропереписи дают большее поле для исследований по сравнению сo всеобщими переписями населения, особое внимание уделено тому, какое отражение итоги микропереписей нашли в научных исследованиях и какие были получены выводы на их основе. &#x0D; Более пристальное внимание уделено микропереписи населения 2015 г. (МПН-2015). Выявлены отклонения выборки МПН-2015 по регионально-территориальной структуре от постоянного населения России. Из-за значительной вариативности объема выборки по регионам пострадала репрезентативность данных по стране в целом, наблюдается смещенность социально-демографических и географических характеристик населения. Сопоставление численности и структуры населения по МПН-2015 с данными текущего учета и всеобщих переписей населения, оценка возрастной аккумуляции показали наличие серьезных проблем с качеством полученных данных, а также наличие типичных для всеобщих переписей ошибок. Акцентируется внимание на смещении в старшие возраста структуры населения по МПН-2015, что может повлиять на оценку общих показателей образования, экономической активности, заболеваемости, инвалидности, демографического состава домохозяйств и других переменных, для которых существует тесная связь с возрастом, что в целом может отразиться на востребованности данных МПН-2015. В заключении, обращаясь к опыту микропереписей и социально-демографических обследований в зарубежных странах, а также выборочным обследованиям Росстата по социально-демографическим проблемам, поставлен вопрос о перспективах развития данного метода статистико-демографического наблюдения в России.