"subsample analysis" is a correct phrase in written English and it can be used when referring to a statistical sample that is a subset of a larger population or dataset
For example, "The subsample analysis indicated that the research subjects in the smaller sample were more likely to have a certain characteristic than the larger population."
The results of subsample analysis for Eagle Ford parallel indicated varying range of pore volume distribution.
Subsample analysis reveals that public sector former workers in education are more prosocial even after controlling for a rich set of characteristics.
Based on this subsample analysis, we construct a disclosure score and investigate whether an effect of the German Accounting Law Modernization Act is only observable for those companies that disclose more information regarding their R&D activities.
After having looked at the public sector as a whole, in this section, we conduct some subsample analysis, disaggregating the public sector into its main industries, looking separately at the most important professional groups, and exploring heterogeneity along the tenure, age, and country dimensions.
For sub-Saharan Africa (SSA), where death rates are highest, we conducted a subsample analysis of motivations for giving birth at home.
For SSA we conducted a subsample analysis of delivery attendance among the aggregated "home" birth respondents using Model Women's Questionnaire question 435 (variable m3a m3m): "who assisted with the delivery of [your child]?" Possible responses include: doctor, nurse/midwife, auxiliary midwife, traditional birth attendant (TBA), relative/friend, other, or no one.
Statistical evaluation of the algorithms was performed by a repeated random subsampling analysis.
Unfamiliar objects, on the other hand, were only reliably decoded in V1 (and on one subsampling analysis in right S2).
Including FDR calculation by subsampling analysis (100 subsamples), the calculation time was 88 hours for DCglob and 116 hours for DCloc.
After generating the initial network, researchers often extract a subsample for analysis (i.e., principal component).
As an example, sampling (5 kg), sample preparation (USDA subsampling mill and 250 g subsample), and analysis (TLC) accounted for 83%, 9%, and 8% of the total aflatoxin testing error, respectively, when testing raw shelled peanuts for aflatoxin.
