Reliability. Marsden (2011, Pp.382-383) Indicates That Respondents Are More Likely to Accurately

Reliability. Marsden (2011, pp.382-383) indicates that respondents are more likely to accurately nominate those individuals with whom they interact regularly and frequently. This would apply to our questions regarding the close colleagues from whom beginning teachers most frequently receive support regarding math instruction. Novice teachers are likely to be less reliable in reporting those from whom they receive infrequent assistance; this is confirmed by Pitts and Spillane (2009). As per Marsden (2011) and Brewer (2000), we will use a roster of school personnel to prompt the beginning teachers on at least one network question. We will also improve reliability by asking novices for multiple network relations (e.g., closest colleagues and those who provide support with math instruction).

Henry, Lubell and McCoy (2012) also discuss prompting people with targeted subsets (e.g. in the marketing department, or in the 2nd grade, who do you know) which can improve reliability.

Marsden also reports that actors are likely to overestimate their out-degree. In our case, this would suggest that mentors, other teacher colleagues, math instructional coaches, and principals might overestimate the extent to which they assist beginning teachers. It is for this reason that we will ask the novice teachers who has supported them, as the recipient of the assistance is the more reliable reporter.

Validity.Pitts and Spillane (2009) validated the social network instruments they used with a population of elementary school teachers very similar to ours. They found 80% agreement between interview and survey responses. Moreover, they found that about 70% of the responses about advice sought for subject specific instruction (e.g., mathematics) reflected interactions intended to influence knowledge or practice or both. That is, teachers were able to separate instructional advice from other forms of information flow or advice. Thus, by asking beginning teachers from whom they receive math instructional advice, we are likely tapping the flows that relate to novice teachers’ math instructional practices as they navigate potentially competing demands in their classrooms.

On the other hand, Pitts and Spillane found that questions about advice sought may not reflect unsolicited advice. But such advice may be very influential for a beginning teacher. Therefore, instead of using Pitts and Spillane’s questions about advice sought, we will ask about advice received, regardless of whether it was solicited.

Brewer, D.D. (2000). Forgetting in the recall-based elicitation of personal and social networks.Social Networks, 22, 29–43.

Henry, Adam Douglas, Mark Lubell, and Michael McCoy (2012). “Survey-Based Measurement of Public Management and Policy Networks.” Journal of Policy Analysis and Management, 31(2): 432-452.

Marsden, P.V. (2011). Survey methods for network data.In J. Scott & P. Carrington (Eds.), The Sage Handbook of Social Network Analysis (pp.370-388). London: Sage Publications.

Pitts, V., & Spillane, J.P. (2009).Using social network methods to study school leadership.International Journal of Research and Method in Education, 32(2), 185-207.

Spillane, J.P., & Hopkins, M. (2013). Organizing for instruction in education systems and school organizations: How the subject matters. Journal of Curriculum Studies, 45(6), 721-747.

For information on general survey work, see

Dillman, Don A.Mail and Internet surveys: The tailored design method--2007 Update with new Internet, visual, and mixed-mode guide. John Wiley & Sons, 2011.