Effects of Internet on Child Development

180 to learn was reported in 65 cases, to play was reported in 57 cases, to browse in 35 cases, and to communicate in 27 cases. Thus, the five indices of child home Internet use in cluded: 1) the continuous variable years of home Internet access and the dichotomous (report ed-unreported) variables of child home In ternet use to 2) learn, 3) play, 4) browse, and 5) communicate. Family Socioeconomic Characteristics The parent questionnaire assessed five family characteris tics commonly used to determine socioeconomic status (Bradley & Corwyn, 2002; Sirin, 2005).Two items queried father’s and mother’s employment status. Approximately 70% of mothers and 96% of fathers were employed, full-time or part-time. Two questionnaire items requested father’s and mother’s level of education, coded as: elementa ry = 1, junior high school = 2, high school incomplete = 3, high school complete = 4, technical school/college (complete or incomplete) = 5 and university (complet e or incomplete) = 6. The mean educational level of mothers was 4. 79 (SD = 0. 95) suggesting that many mothers had post-secondary education; the mean educational level of fa thers was 4. 45 (SD = 1. 2) suggesting that some fathers had post-secondary education. The final socioeconomic item on the questionnaire asked parents to indicate annual family income by selecting one of the following options: < $20 000 = 1, $20 000 to $40 000 = 2, $40 000 to $60 000 = 3, $60 000 to $80 000 = 4, $80 000 to $100 000 = 5, > $100 000 = 6. Annual income for participating families was approximately $60,000 CD (M = 4. 07, SD = 1. 48). Table 2 presents a summary of measured constructs which includes: four tests of children’s cognitive development, five indices of children’s home Internet use, and five fa ily socioeconomic characteris tics. Which are the better predictors of cognitive development during childhood, — el ements of the microsystem or elements of the techno- subsystem? Two series of stepwise regression analysis we re conducted with the four c ognitive development scores as the dependant variables. In the first regression analyses , family socioeconomic characteristics (elements of the microsystem) were the independent variables. In the second analyses, indices of home Internet use (elements of the techno-subsystem) were the independent variables. Tab le 2 Description of Constructs and Measures Ecological System System Elements Specific Measures Bioecology Cognitive Development Expressive Language Metacognitive Planning Visual Perception Auditory Memory Techno-Subsystem Home Internet Use Years of Internet Access Online Learning Online Playing Online Browsing Online Communication Microsystem Family Characteristics Father Employment Mother Employment Father Education Mother Education Annual Family Income Results Results of analyses revealed that fa mily socioeconomic characteristics (eleme nts of the microsystem) explained a odest (but significant) amount of the variation in children’s cognitive deve lopment scores. As presented in Table 3, adjusted R 2 values indicated that father’s level of education accounted for approximately 7% of the variation in children’s level of expressive language (as measured by the WISC-IV vocabulary subtest), 5% of the variation in children’s visual perception and auditory memory (as measured by the CAS nonverbal matrices subtest and CAS 181 word series subtest, respectively). Whether or not moth ers were employed, part-tim e or full-time, accounted for pproximately 6% of the differences in children’s capacity to execute metacognitive functions such as planning (as measured by the CAS matching numbers subtest). While the other measures of familial socioeconomic status (e. g. , mother’s education and family income) explained some of the variance in children’s cognitive development, such measures did not improve upon the predictive utility of fa ther ’s education or maternal employment; variation is prerequisite to prediction. Almost all fathers were employed and almost all mothers had finished high school. For participating middle-class families, father’s education a d mother’s employment were more sensitive to children’s cognitive development scores than were family income, father’s employment, and mother’s education. Tab le 3 . Stepwise Regression Analysis: Family Characteristics Predicting Child Cognitive Development Cognitive Score Predictor Beta Weight t value R 2 (adj) F value Expressive Language Father Education . 292 2. 70** . 074 (1, 78) = 7. 29** Metacognitive Planning Mother Employed . 270 2. 46* . 061 (1, 77) = 6. 05* Visual Perception Father Education . 244 2. 22* . 047 (1, 78) = 4. 93* Auditory Memory Father Education . 258 2. 6* . 054 (1, 78) = 5. 55* *p < . 05; **p < . 01 Results of analyses further revealed th at indices of home Internet use (elements of the techno-subs ystem), in general, explained more of the variation in children’s cognitive de velopment than did family socioeconomic characteristics (elements of the microsystem). Summarized in Table 4, specific types on online behavior (i. e. , learning, communicating, and playing) and years of home In ternet access combined to predicted child cognitive developmental outcomes. Indicated by adjusted R 2 , children’s online communication, ears of home Internet access, and online learning (as reported by parents) accounted for ap proximately 29% of the varia tion in children’s level of expressive language as measured by the WISC-IV vocabulary subtest. Online learning and communicating (reported- unreported) combined to explain 13. 5% of the variation in children’s metacognitive planning. Online learning and playing (reported-unreported) combined to explain 10. 9% of the variation in children’s auditory memory. Years of home Internet access explained approximately 3% of the diffe rences in children’s visual perception scores. With the xception of visual perception, indices of home Internet use (elements of the techno-subsystem) were better predictors of children’s cognitive development than were family socioeconomic characteristics (elements of the microsystem). Tab le 4 . Stepwise Regression Analysis: Home Internet Use Predicting Child Cognitive Development Cognitive Score Predictor/s Beta Weight t value R 2 (adj) F value Expressive Language Online Communication . 344 4. 00*** Years of Internet Access . 263 3. 12 ** Online Learning . 256 2. 99** . 287 (3, 101) = 14. 97*** Metacognitive Planning Online Learning . 287 3. 03** Online Communication . 201 2. 12* . 35 (2, 101) = 9. 06*** Visual Perception Years of Internet A ccess . 192 1. 99* . 028 (1, 104) = 3. 98* Auditory Memory Online Learning . 242 2. 60* Online Playing . 228 2. 46* . 109 (3, 101) = 14. 97*** *p < . 05; **p < . 01; ***p < . 001 Discussion A variety of mechanism s linking family socioeconomic status to child cognitive development have been proposed including parenting (Petrill, Pike, Price, & Plomin, 2004 ; Mistry, Biesanz, Chien, Howes, & Benner, 2008) and 182 resources (Bradley & Corwyn, 2002). For the current samp le of middle class children, paternal education and maternal employment were associated with measures of hild cognitive development. More educated fathers tended to have offspring who scored high on three of the four cognitive measures (expressive language, visual perception, and auditory memory). Mothers who were employed tended to have children who scored high on the measure of metacognitive planning. Educated fathers and employed mothers may genetically transmit to their offspring some neurological processing advantage (bioecology). Simultaneously, educated fathers may provide enhanced language models and stimulating environments that facilitate the cognitive development of their children (microsystemic influence). Employed mother may provide models of organization and place increased demands on children to self- regulate thereby enhancing the metacognitive planning abilities of their offspring (microsystemic influence). Family socioeconomic status (as measur ed and for the current sample) accounted for 5% to 7% of differences in child cognitive development scores. In contrast, indices of home Internet use (as measured and for the current sample) accounted for 3% to 29% of differences in child cognitive development scores. Me ta-analysis confirms that the impact of socioeconomic status on academic achie vement is eroding over time (Sirin, 2005). Increasingly ffective structures of social equali zation (e. g. , public education, quality daycare, preschool intervention, and prenatal programs) and the expanding middle class create the need for more precise description of home environments. Current results suggest th at indices of home Internet use (i. e. , elements of the ecological techno- subsystem) pro vide more useful information regarding cognitive development than do family socioeconomic characteristics (elements of the microsystem). Only two of five family socioeconom ic characteristics added to the regres sion equation, suggesting that some measures (i. e. , family income father employment, and mother education) did not differ in relation to children’s cognitive development. In contrast, four of the five indices of home Internet use during childhood added to the regression equation, suggesting that these measures differe d in relation to children’s cognitive development. In the context of the current investigation, soci oeconomic status is a crude construct re lative to home Internet use. Internet use includes both organized (e. g. , search) and disorganized (e. g. , browse) interactions with both human (e. g. , chat) and nonhuman (e. g. , database) elements in online environments (Johnson & Kulpa, 2007).Internet use is a complex set of behaviors that vary widel y across individuals and th at is influenced by cognitive and personality characteristics (Joinson, 2003). For the current sample of children, patterns of home Internet use explained more of the variation in cognitive development than did family socioeconomic characteristics. In the context of middle class families, elements in the techno-subsystem (e. g. , Internet access) may not necessarily facilitate child cognitive development; effective use of those elements, highly dependent upon parent behavior, may promote development.For example, Cho and Cheon (2005) surveyed families and found that parents’ perceived control, obtained through shared web activities and family cohesion, reduced children’s exposure to negative Internet content. Lee and Chae (2007) reported a positive relations hip between parental mediation techniques (website recommendation and Internet co-use) and children’s educa tional attainment. In the current investigation, the cognitive experienc es provided to children by employed moth ers may include Internet skills instruction (e. g. , sending email) and models of information management (e. g. acc essing websites for informa tion). Such experiences, over time, may provide children with enhanced opportunities to direct their own cognitive development via increasingly sophisticated uses of the Internet. According to Livingston and Bober (2005), â€Å"a new divide is opening up between those for whom the internet is an increasingly rich, diverse, engaging and stimulating resource and those for whom it remains a narrow, unengaging, if occasionally useful , resource of rather less significance† (p. 2). Bruner (2005) recen tly reiterated that â€Å"our minds ap propriate ways of representing th world from using and relating to the codes or rules of available technology† (p. x). Cognitive abilities prerequisite to utilization of Internet applications constitute an implicit component of contemporary notions of intel ligence (Maynard, Subrahmanyam, & Greenfield, 2005). The ecological techno-s ubsystem furthers our understanding of environmental influences on child development by emphasizing the impact of digital technologies on cognitive growth during childhood. The techno- subsystem provides precise description of microsystemic mechanisms of developmental influence which lead to intervention strategies.According to Livingston and Bober ( 2005), many parents lack the skills to guide and support their children’s Internet use and Intern et-literate parents have Internet-litera te children. Subsequent research may evaluate the effectiveness of techno-subs ystem interventions for elementary school children at-risk, for example, the provision of home Internet access and pa rent Internet literacy training. As stated elsewhere, â€Å"current anxiety surrounding children’s Internet use should be for those whose cognitive processes are not influenced by the cultural tool† (Johnson, 2 006, p. 570).