This experiment examined the impact of pre-merger identification (low, high), pre-merger group status (low, high) and relative representation (low, high) upon identification with a new merger group. In the first phase of the study, 156 university students were assigned to a pre-merger team of ‘inductive’ thinkers. Pre-merger identification was manipulated by giving the participants feedback about the extent to which they used the inductionist style. Pre-merger status was manipulated by informing participants that their team had performed worse or better than a deductionist team on a decision-making task. In the second phase, the two pre-merger teams were combined into a merger team of analyst thinkers. Relative representation was manipulated by maintaining most or none of the features of the pre-merger team in this new merger team. The results revealed that high pre-merger identifiers identified more strongly with the merger group than did low pre-merger identifiers, but only when the relative representation was high. Pre-merger status did not influence post-merger identification. Copyright © 2006 John Wiley & Sons, Ltd.