Purpose: Analysis of children's oral narratives is a frequently used naturalistic assessment technique. Comparing children's oral narrative performance to databases of samples elicited from typically developing speakers aids in the identification of language impairment and thus enhances the clinical utility of the assessment process. To investigate the potential usefulness of existing databases across different geographic locations, this study compared the story retelling performance of English-speaking children from New Zealand (NZ) to samples from the United States (US) contained in a widely used reference database available with Systematic Analysis of Language Transcripts software (SALT). Method: Sixty-six NZ children (age 6;0 to 7;11) who showed typical development participated. Their performance was compared to 73 age-matched samples from the United States. All children retold the story Frog, Where Are You? using a standard protocol. Approximately half of the NZ children (n = 31) retold the story without picture support, whereas all other children were allowed to refer to the pictures during retelling. Language samples were analysed on measures of verbal productivity, semantic diversity, syntactic complexity, verbal fluency, and story quality. Results: The results indicated that variables measuring verbal productivity, semantic diversity, and story quality were sensitive to changes in elicitation procedures (presence or absence of pictures during retelling), but not to differences in geographic location (US vs. NZ). In contrast, verbal fluency was sensitive to both elicitation condition and geographic location. Implications: The results from this study suggest that, when comparing a story retelling sample to a reference database, adhering to the language sampling elicitation protocol may be more important than the geographic origin of the database.