Working memory is the temporary storage mechanism that underlies our ability to hold and process information over a brief period of time. It is a pre-requisite for the long-term storage of information, reading, problem-solving and language comprehension. Research examining the relationship between working memory and socioeconomic status remains inconclusive. The present study brings together educational, cognitive ageing, intelligence and socioeconomic status (SES) research in order to determine whether, in a large, adult sample, working memory performance differs across SES. Data were obtained form an Internet test attempted by ~480,000 people on the Science pages of the British Broadcasting Corporation’s (BBC) website. This data included tests of working memory, short-term memory and some demographic information on participants. A subset of participants who provided the necessary demographic information and were aged between 30 and 45 were selected. Participants’ postcodes were linked to Geoplan® (2011) SES data, and working memory performance was assessed across SES. Tests of measurement invariance identified a single latent working memory factor that was being measured consistently across SES. It was not possible to constrain means equal. Therefore differences in mean performances across SES could be considered to reflect differences in the latent working memory construct. Factor scores on this latent variable were computed and used as a measure of participants’ general working memory performance. A series of ANOVAs demonstrated that working memory performance differed from SES group AB to C1 (p=0.0000035), and from AB to D (p=0.0000024). When education was introduced as a covariate, there was a significant difference between the working memory performance of participants in SES groups AB and C1 (p=0.004), AB and D (p=0.003), and C1and D (p=0.012). Performances on individual working memory tasks were either significant or almost significant in comparisons between AB, C1 and D. In all analyses, there were no significant differences observed when SES grade E was included in pairwise comparisons. Results indicated that SES influences level of working memory performance, with those in a higher SES grade scoring higher on average than their lower SES counterparts. Implications for further research are discussed with a focus on SES measurement and conception, as well the problems associated with demographic composition of the present sample.