Social Network Blog - y1

Why do you think that comparing methods has been more popular than having methods complement each other?

This is a very good question. I think that researchers are so keen on putting their own research method on a pedestal that they completely ignore the possibility of using several measurement methods to find a more accurate measurement. My guess is that this has to do with a few factors: first, researchers are worried that using complimentary methods, in addition to their own, might overshadow their findings/ analysis. Another possible reason would be accessibility—can researchers access the right measurement techniques from other researchers?

Your question (of what I would predict the results of the Lin study to be if it were done in the United States) made me think of Smith-Lovin. Her study showed how core discussion groups shrank from 3 to 2 between 1985 and 2004. When analyzing the results by gender, she determined that females had expanded their contacts to include people outside of the home, whereas men were discussing important matters more exclusively with kin. Such a shift is indicative of the changing, dual role of women; such a fact is more intriguing when juxtaposed against the contents of men’s core discussion networks.

More importantly, such a situation applies here. If this study were done in the United States in 1985, the results would likely be more similar to Lin’s results from Taiwan than if this study were done in the United States in 2004. This presumptuous statement is based largely on Smith-Lovin’s findings. Even so, I do predict that women in the United States would have higher social capital – as defined by occupational contacts – than the women in Taiwan. In addition, I think that the women in the United States would still have more community-based social contacts, on average, than men in the United States (and men in Taiwan).

I definitely think it would be very interesting to replicate this study in the United States (as both a measure of the types of social contacts people have as well as an evaluation of the Position Generator methodology).

If the Resource Generator yields the same kind of results as the name generator and the position generator, and also have some of the same methodological problems, such as biasing/influencing the responses to the questions depending on the wording (15), then how is the resource generator a better measure of social capital compared to the other measures?

The resource generator according to Van der Gaag & Snijders is a survey instrument for the measure of social capital. Like you said, it is similar to the position and name generator. However, it differs in its ability to directly refer to social resources instead of occupational prestige in the position generator and thus is more broad and more applicable than the other methods. Also, certain methods are better for certain types of studies, for certain types of research questions. In some cases, combining methods may not actually help answer the question.

Why do you think that comparing methods has been more popular than having methods complement each other? In what circumstances would contribution of methods be better than the comparison of techniques?

I think that combining methods create a stronger study in terms of reliability and validity, but maybe not for feasibility. A lot of the problems with the more extensive studies is that time-consuming, burdening activities generally don’t get a lot of participation and are generally not as applicable in say, organizational settings.

I personally believe that the reason the articles, most explicitly the DeJong article, likes to contrast different studies is because the difficulty in defining social capital and properly mapping communication across entire social networks is extremely immense. As a result it is easy to find flaws and contradictions between studies. None of the studies have been able to perfectly or even close to perfectly map social networks and use this information to apply across different social groups. However, I think you question is valid and I have to assume a level of pride by the author to prove ones own point also plays a roll, but a combination of many different data collection methods working together to explain the findings can only help make the results more accurate and universal.
The Lin article, I thought was biased by the social structure of the subjects that were chosen for the data collection. I think and hope that the results would be significantly different if were recorded on the social capital of women vs men in the untied states. I believe that women in the US are born with an inherent level of social capital, based on their significance as mother and care taker in past generations. I purposed that women have higher levels of social capital among extremely weak or complete stranger. Are you not more likely to walk a woman across the street than a man, or help a woman move furniture than a man? In the article Lin suggests that the position of care taker by women gives them a level of social capital not replicated in men’s social structures. Men are forced to gain most of their social capital from labor connections. The articles do not go as far as suggesting which social capital results in a happier life, but I would argue that simply being a woman allows for a higher level of social capital than men, and social capital that is more valuable than that of men.

“If the study were to be done in the US, in what ways/dimensions do you think that the results would differ to those found by Lin et al?”

This question reminded me of the Bott study and the two McPherson et al studies on core discussion groups and homophily. Based on their results it would be safe to say that if the Lin et al study were replicated in the US, the results would be quite similar, yet the degree of inequality would be smaller between men and women. For instance, men would still have an advantage of being in the labor force, compared to women who are likely to stay at home. However, as more and more women join the workforce in the US, there would be less of a difference. Men might also have more social capital in the form of non-kin ties when compared to women. However, we saw that men are losing their non-kin ties and this might close the gap between men and women’s social capital status. Their findings on the return of access to social capital would be quite similar as well, as men are more likely to benefit from social capital when it comes to getting higher prestige jobs and higher incomes. This might be explained through the importance of homophily. Men in higher positions would be more likely to get along with other men, rather than women.

Why do you think that comparing methods has been more popular than having methods complement each other? In what circumstances would contribution of methods be better than the comparison of techniques?

I think that studies chose to focus on a comparison of existing methods because it is an excellent way of testing the reliability and validity of existing methodologies. This provides valuable information concerning the body of research in the field of study because it sheds light on which studies provide findings and observations that are statisicallty valid and can be cited in other areas of research. In order to find new methodologies, it is important to determine what measures have worked in the past, and which ones have not. Reseachers may chose to focus on one methodology rather than a combination of several due to the constraints of the research, such as time, money and other varying factors.

This question relates to the important part of experimental research cocerning the determination of causality. In order to make claims of a causal relationship, it is important to rule out alternative explanations.