Classic study: van IJzendoorn and Kroonenberg (1988)

Aims and procedure

Aims: to compare attachment types within and between countries, and to use several samples from the same countries to reduce the bias of any single sample.

Procedure: a meta-analysis. A database search found Strange Situation studies, which were included only if they:

• studied infant-mother attachment with the Strange Situation and gave A, B, C classifications;
• were not special samples (for example, children with Down's syndrome);
• had 35 or more infants, with no overlapping samples, and infants under two years old.

Ainsworth et al. (1978) was used as the planned baseline. The final data: 32 samples from 8 countries, 1,990 Strange Situations.

Results

Totals by country computed from the study-by-study table in Pearson's WPS03 summary of studies. The booklet's 'remaining 17 studies' row is counted as USA: the original paper had 18 US samples totalling 1,230 infants (105 + 1,125)

Findings and conclusions

• Type B (secure) was the most common type in every country: 65% overall, then A (21%) and C (14%).
• Western European countries (such as Germany) had more type A and fewer type C than expected; Israel and Japan had more type C and fewer type A. The USA sat between.
• Variation within countries was nearly 1.5 times the variation between countries, and was especially large in Germany and the USA.
• The combined US studies did not match Ainsworth's distribution, so there was no agreed baseline.
• The authors suggested (speculatively) that mass media spreading similar child-rearing ideas might explain the global dominance of type B.

Evaluation

• Strengths: a very large combined sample increases generalisability and reliability; strict inclusion criteria; within a country the same researchers often ran the studies, so procedural differences are unlikely to explain within-country variation.
• Weaknesses: some samples were small, so one study cannot represent a whole culture; the 32 samples are unevenly spread (most from the USA; only one each from China, Great Britain and Sweden); 'country' is not the same as 'culture' (within-country variation was large); the Strange Situation itself may be an imposed etic; the media explanation was not tested.

Worked example

Using the totals for Japan, calculate the percentage of resistant (type C) infants and compare it with the overall figure.

1. Japan: n = 96, type C = 26.
2. 26 / 96 x 100 = 27.1%.
3. Overall type C = 273 / 1990 x 100 = 13.7%, so Japan has about double the overall rate.

Answer: 27.1% in Japan versus 13.7% overall.

Check yourself

1. What kind of study is van IJzendoorn and Kroonenberg (1988)?
   • A new Strange Situation experiment in eight countries
   • A longitudinal case study
   • A questionnaire survey of mothers
   • A meta-analysis of existing Strange Situation studies
   Show answer

   A meta-analysis of existing Strange Situation studies. It pooled 32 existing samples.

2. Overall, 1294 of the 1990 infants were type B. What percentage is that? Give your answer to 1 decimal place.
   Show answer

   65.0 %. 1294 / 1990 x 100 = 65.0%.

3. In the German samples combined, 48 of 136 infants were type A. What percentage is that? Give your answer to 1 decimal place.
   Show answer

   35.3 %. 48 / 136 x 100 = 35.3%.

4. What did the study find about variation within and between countries?
   • Variation within countries was nearly 1.5 times greater than between countries
   • Countries were all identical
   • Variation between countries was much greater than within
   • There was no variation at all
   Show answer

   Variation within countries was nearly 1.5 times greater than between countries. Within-country differences were larger.

5. Which was an inclusion criterion?
   • Only studies of fathers
   • Only children with Down's syndrome
   • Samples of 35 or more infants
   • Infants aged over two years
   Show answer

   Samples of 35 or more infants. Samples under 35 were excluded.

6. Which countries showed relatively more type C (resistant) attachment?
   • Israel and Japan
   • Germany and Great Britain
   • Sweden and the Netherlands
   • The USA and Germany
   Show answer

   Israel and Japan. Type C was more frequent in Israel and Japan.

7. Explain one weakness of van IJzendoorn and Kroonenberg's (1988) meta-analysis.
   Show answer

   The samples were unevenly spread, with most from the USA and only one study from China, so the results for some countries may not represent their whole culture.. Name the problem and its effect.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification, Issue 6 (November 2024, first assessment January 2026)
• Pearson IAL Psychology summary of studies: WPS03
• van IJzendoorn and Kroonenberg (1988), Cross-cultural patterns of attachment: a meta-analysis of the Strange Situation, Child Development 59 (Leiden University repository)

Last updated 8 November 2026. Spotted a mistake? Tell us.