Understanding the world requires unbiased and valid empirical knowledge. Numerous studies on the same topic, employing different survey data, often produce divergent analytical results and even contradictory conclusions, which undermines the effective testing of theoretical reliability and applicability. However, existing studies predominantly focus on refining statistical methods while overlooking foundational issues such as sample representativeness. There is also a scarcity of systematic examinations into the sample structures of widely used large-scale social surveys and their impact on statistical findings.

To address this gap, this study draws on six most extensively used national large-scale social surveys among Chinese scholars. It compares their sampling designs and empirically investigates the similarities and differences in their sample structures. Using a consistent model specification, this study investigates the impact of deviations in sample structure on statistical analysis results, and reveals the underlying logic by which sample structure influences statistical inference.

The main findings are as follows. First, although almost all surveys employ a multi-stage, stratified Probability Proportional to Size (PPS) random sampling method, they exhibit significant differences in sampling frame coverage, stratification principles, the sampling methods and quantities of sampling units at each stage, and within-household sampling procedures. Second, notable disparities exist in the distributions of key demographic variables across the surveys. Moreover, each survey's sample structure deviates to some extent from that of the 2015 National 1% Population Sample Survey. Third, differences in sample structure lead to variations in statistical results. Under identical models, analyses based on different survey data yield both a consensus component reflecting shared social realities and significant discrepancies in the significance and direction of effects for certain variables. Fourth, adjustments in population definitions, weighting schemes, variable selection, and operationalization alter the joint distribution of variables within a sample, thereby significantly affecting statistical outcomes. When sample structures differ initially, such adjustments may further amplify discrepancies in results across different survey datasets. Fifth, the foundational role of sample structure in the methodology of statistical inference must be fully acknowledged.

Based on these findings, the study recommends that researchers should meticulously review survey technical documentation,prudently select appropriate survey data based on research objectives, appropriately address data missingness and weighting, prioritize robustness checks of analytical results, and thoroughly evaluate or explain the sample representativeness of the survey data used. Survey institutions, on the other hand, should provide more detailed weighting information and comprehensive technical documentation to enable researchers to use the data more appropriately.

The primary contributions of this study are as follows. (1) It employs empirical methods to systematically examine the sample structures of six large-scale social surveys and the impact of sample structure deviations on statistical results, revealing methodological pitfalls that offer a new perspective for understanding the contradictory conclusions drawn from different datasets in existing literature. (2) Theoretically, it extends methodological reflection in quantitative research from model specification back to the data-collection stage, broadening scholarly discourse. (3) Practically, it provides empirical guidance for standardizing data usage in quantitative research, thereby enhancing the comparability and robustness of research conclusions.

Under the background of negative population growth, whether the scale of China's floating population will continue to grow and how its structure will change in the future are related to the construction goals of relevant social systems. Based on the data of the fifth, sixth, and seventh national population censuses, age structure standardization and decomposition methods are used to analyze the relative role of structure effect, scale effect, and intensity effect in the change of China's floating population in the past 20 years, and population projections are used to examine the future trend of the floating population. Research has found that the increase in the size of China's floating population in the past 20 years is mainly due to scale and intensity effects. Among them, the increase in the size of the urban floating population is mainly due to scale effects, while the increase in the rural floating population is mainly due to intensity effects; The structure effect has always had a negative impact, with the negative effect in the last 10 years being greater than that in the first 10 years. This negative structure effect shows similar behavior among different types of floating populations. Assuming the age-specific migration rate remains unchanged, the size of China's floating population will continue to decline in the future, and its age and urban-rural structure will undergo significant changes. The structure effect can be regarded as an intrinsic factor of natural growth, and the study of China's floating population must be situated within the context of negative population growth and population ageing.

Birthplace information is crucial for migration studies, enabling identification of lifetime migrants. Using data from 2000, 2010, and 2020 censuses, this research depicts inter-provincial lifetime migration in China, outlines national- and provincial-level migration trends, directions, and age patterns. The results suggest that lifetime migration reflects historical and contemporary migration, while the historical marks gradually fade. Its spatial distribution shows the characteristics of diffusion for the places of origin and diffusion-and-concentration for the destinations, while China's population remains relatively static. Cohort analysis highlights that lifetime migration has a cumulative effect over time and is closely associated with life events. The study concludes that birthplace information has a unique research perspective distinct from others, and calls for further research on the conditions and causal mechanisms of bidirectional migration flows.

Based on the 2017 China Migrants Dynamic Survey and data from relevant statistical yearbooks, this paper uses nested logit model (NLGT) to analyse the destination selection mechanism of floating population in China. This study takes prefecture regions as the basic geographical analysis unit and the flow sample as the research object. The results show significant structural differences between flow sample and stock sample, and heterogeneous effects of the same variables on different samples. The socioeconomic returns that floating population can personally feel play an important role in the destination selection mechanism. The interaction between regional level and individual characteristics shows that regional level characteristics have a heterogeneous effect on the selection mechanism. The paper suggests that focusing on flow or stock samples should be based on the specific study aim of different research questions. Expected socioeconomic return, rather than expected income, is one of the most important factors attracting floating population. The destination selection is a rational and comprehensive decision made by floating population based on individual characteristics and the regional level characteristics of the destination.

Based on data of China Migrants Dynamic Survey (CMDS) conducted in the origin and destination in 2015, this paper compares the structural differences of the floating population under the two survey perspectives, with the mini-census in 2015 as the benchmark. The results suggest that the survey conducted in origin is closer to the results of the mini-census in age structure, marriage, and distance of movement. The survey conducted in destination is closer to the mini-census in terms of residence time. Women in the destination areas and short-term (1 year or less) migrants in the origin areas are over-sampled. Both surveys focus on the population of low education level while neglecting the population with college degree or above. From the perspective of survey methodology, recruiting different respondents is one of the most important sources of survey errors, especially measurement errors. In addition to coverage error, non-response error is also one of the important factors affecting the survey results and sample structures.

 

With the counter-factual causal inference framework，this paper employs various propensity score matching methods to explore the effect of selectivity of migration on the fertility level of floating population in China，and estimates and compares migrant and non-migrant fertility levels after controlling for the selectivity bias. Both lifetime fertility( children ever born) and period fertility( births in the year proceeding the census) are examined using data from the 2000 census. Heckman 2-stage model shows that the observed difference between migrant and non-migrant fertility is biased due to selectivity of migration. The consistent results from three types of propensity score matching methods demonstrate that migrant fertility is markedly lower than non-migrant fertility，and a larger proportion of the difference is explained by the selectivity. Anyway，the study suggests that migration does reduce fertility.

After summarizing the theories and measurement indices of assimilation,this paper examines and re-constructs the measurement framework of assimilation at individual level,followed by discussing some theoretical questions of assimilation in present China.Assimilation could be one point on the line connecting adaptation,segmented assimilation and assimilation.Measurement indices framework of immigrant assimilation in destination should be simplified with high validity,corresponding to the theories.Assimilation can be divided into five dimensions: economic,cultural,social,structural and identity.Future research should be directed at the final direction of the assimilation of migrants,and the causal chain centered on the assimilation(the status and the causes and consequences of assimilations),and longitudinal or panel survey studies should be enhanced to understand the situation and development trajectory of assimilation of migrants in China.

This paper examines the psychological status and its development trajectory of migrant children based on three-wave survey data of "Panel Study on the Development of Migrant Children"(PSDMC).Despite the fact that psychological status has been improving for all three groups of children,when measured by loneliness and depression,psychological status of migrant children is worse than that of resident children.The development trajectory of migrant children who study in public schools is the same as that of the resident children,while it is vastly different from that of migrant children who study in migrant schools.And as one aspect of psychological status stemming from social inequality,depression of migrant children studying in migrant schools has not been improved in the three waves of PSDMC.