## Glossary

- SACMEQThe Southern and Eastern Africa Consortium for Monitoring Educational Quality
- samplea subset of a target population that can be studied for purposes of making generalizations about the larger target population
- sample designall aspects and features of a specific sample, such as multiple stages or phases, stratification, and clustering
- sample sizedetermines the number of units to be sampled
- sampling framea list of all units belonging to the target population, or in the case of multiple-stage sampling, a list of all units containing the target population
- sampling weighta numeric value representing the inverse of the selection probability; in an ILSA, the final sampling weight (also called “design weight” or “estimation weight”) usually reflects all sampling stages and includes non-response adjustments
- SAQsurvey activities questionnaire
- SASinitially short for “Statistical Analysis System”, analytical software produced by SAS Institute Inc.
- scalemetric used to summarize performance on a test or on a set of items in a background questionnaire
- scale anchoring procesused to arrive at probabilistic descriptions of what study participants (e.g., students) know and can do at different levels of a distribution of scores
- scalingprocess of creating scales
- SCCSEACMEQ (previously SACMEQ) Co-ordinating Centre
- SEACMEQThe Southern and Eastern Africa Consortium for Monitoring Educational Quality (previously SACMEQ)
- secondary analysesanalyses performed on data collected by others; these generally follow the primary analysis performed by the organization or agency responsible for collecting the data
- SESsocio-economic status
- significancethe level of probability reflecting the maximum risk one is willing to accept that observed differences are due to chance; typically set at .01 (a 1% chance that a sample statistic will be due to chance) or .05 (5% chance)
- simple random samplingsingle-step method to ensure that every possible sample of size n has an equal chance of being selected, thus ensuring that each each unit in a sample has the same inclusion probability
- SMIRCScience and Mathematics Item Review Committee
- SPSSStatistical Package for the Social Sciences (now “IBM SPSS Statistics”), a software package presently distributed by IBM
- SRSsimple random sampling
- STEMscience, technology, engineering, and mathematics
- STEPSkills Towards Employment and Productivity; a study conducted by the World Bank
- stratified samplingprobability sampling technique by which a sampling frame is divided into homogeneous, mutually exclusive groups (strata) based on specific characteristics (e.g., region, gender), from which independent samples can then be drawn; stratification can be applied at each sampling stage

- SYSsystematic sampling
- systematic samplinga probability sampling procedure according to which units are selected from a population at regular intervals, beginning from a random starting point, until the desired sample size is reached