Data Type         Strengths             Weaknesses                     Source(s) Examples/references
Simple traits, based on expert natural historian knowledge and opinion. Easy to measure. Available for most widely-known species. Difficult to verify. Often missing measurement types (e.g., mean, median or mode). Usually lack measures of variation. Terminology and meaning is often taxon-specific. Field guides, books. Typically lifespan, age-at-maturity, gestation interval, clutch size, frequency of reproduction, mass at birth and at maturity.
Simple traits, derived from published measurements. Verifiable using primary literature. Often available in open databases. Often include population-level replicates or estimates of variation. Available for fewer species, especially in more charismatic taxa. Verifiable sources may be difficult to find. Primary literature, open databases. Databases containing many thousands of species for mammals, fish, reptiles, birds, amphibians, flowering plants [13,15–17].
Life cycle models with age- or stage-based schedules of survival (lx) and reproduction (mx), e.g., life tables, projection matrices, integral projection models. Quantifies whole lifespan Popular for both plants and animals Verifiable using primary literature Often available in open databases Often include population-level replicates or estimates of variation Large toolbox of methods to derive diverse life history trait measurements Can generate derived life history traits (see below). Not available for most species, mostly concerns tetrapods with many broad taxa neglected. Data labour-intensive to collect Often synthesised from multiple sources (sometimes even interspecific). Vital rates measured with variable precision and often contain errors in inference or parameterisation. Vary in length / dimension. Require expertise to handle data and calculate derived measures (usually programming). Primary literature, open databases. Databases available for plants and animals [10,11,18], and detailed data for humans [63].
Life history traits derived from life cycle models using algebraic and computational methods. Benefit from all advantages of life cycle models as above Overcome the issue that models vary in length / dimension. Standardised sets of measurements amenable to comparative analysis. Measures include entire life cycle Measures include many which are not observable (e.g., life expectancy, generation time). Suffer from disadvantages of life cycle models as above: taxonomic breath, data requirements and sources, measurement error. Often assume conditions not met in real systems, e.g., density-independent population growth, stable age/stage structure. Possible to conflate life-history and demographic traits derived from models, e.g., asymptotic or transient population growth. Derivation from life cycle models. [39]
Selection pressures on traits, describing the “importance” of vital rates to fitness using the derivative of the latter with respect to the former, e.g., elasticity or sensitivity. Infer the filter that converts vital rates into fitness and imposes selective outcomes. Provide well-established framework for life history variation in plants. Selection pressures are not life history traits per se. Often assume the same conditions as life cycle models as above. Elasticities usually have sum-constraints across vital rates, hence life-history trade-offs are inevitable. Derivation from life cycle models. [41]