Interpreting a survey on language attitudes in a Polish-Australian community
Your class reads a small survey of 60 Polish-background families in Adelaide. Results:
- 78% of parents say Polish is 'very important' to family identity.
- 41% of children aged 10-15 report speaking mostly English with siblings.
- 62% of parents say they 'always' speak Polish to children.
- Only 29% of children say their parents 'always' speak Polish to them.
A classmate concludes: 'Parents are lying.' Another says: 'The data prove Polish is dying.' Your teacher asks for a more careful interpretation, noting the sample is small and self-selected.
1. Explain two methodological and two interpretive reasons the 'parents are lying' conclusion is unjustified.
Read the worked answer
Methodological reasons:
- Self-selection and small n. 60 self-selected families is not representative. Families who respond to a language survey are likely those most engaged with Polish - inflating the 'very important' figure.
- Different question frames. Parents were asked about their own behaviour ('do you speak Polish to children'); children were asked about their perception of parental behaviour. These are not the same measurement, so a gap is not automatically a contradiction.
Interpretive reasons:
- Social desirability bias. Parents may over-report Polish use because the survey signals that Polish is valued. This is a reporting bias, not deliberate dishonesty.
- Code-switching and domain effects. 'Always' is ambiguous. A parent may speak Polish for instructions and English for homework help, and honestly report 'always' meaning 'as the main home language'. Children may notice the English portions more.
'Lying' attributes intent; the data support at most a reporting gap shaped by question design and norms.
2. Evaluate the claim 'the data prove Polish is dying' and propose what the data actually support.
Read the worked answer
'Dying' is too strong and the data cannot prove it. What the data show is a shift in sibling domains: 41% of children mostly use English with siblings. Sibling talk is often the first domain to shift in migrant communities, because siblings share the school and peer world where the majority language dominates. That is a real signal of intergenerational transmission pressure - but it is not language death.
Three cautions:
- Sample. 60 self-selected families cannot support a population claim.
- Domain specificity. The survey does not measure Polish use in church, Saturday school, grandparents, media, or visits to Poland. These domains can sustain the language.
- Attitudes vs practice. 78% valuing Polish as identity is a resource, not an epiphenomenon. Identity value can motivate maintenance even when daily use declines.
What the data do support: a domain-based shift in child-to-child communication, and a gap between parental report and child perception. A defensible conclusion is 'Polish is under pressure in peer domains, with maintenance resources in identity attitudes - warranting a larger, domain-stratified study'. That is the level of claim the evidence can bear.