Dave's ESL Cafe's Student Discussion Forums

[tisogai]

Please help me figure out a couple of parts in the context below.

Participants

Chapter 1 provides a detailed description of the study design and participants. At 4 2/1 years, 1,083 children and their parents were still enrolled in the study. Mothers had an average of 14.4 years of education, and 16.5% were single. Average family income was 3.6 times the poverty threshold; 79% of the infants were non-Hispanic European Americans. The participants differed in the following ways from the 281 children who were recruited but were lost during follow-up: Mothers of participants had significantly
(p< .05) more education (M=14.4 years vs. 13.6 years) and higher family incomes (income-poverty ratio: M=3.6 vs 3.2) and were more likely to have a husband or partner in the household (85% vs 76%).....

Re the first underline, is it talking about the other different group from the participants? And would you please rephrase for me to understand??

Re the second underline, what does it mean?

Re the third underline, does M mean Mothers?

Re the fourth underline, what does it mean?

Thanks so much.

[CP]

There were 1,083 children and parents still enrolled in the study after 4-1/2 years. An additional 281 children had started out in the study but dropped out. The ones who dropped out were much different from the ones who stayed in the study.

The mothers had more eduction, with an average (arithmetic mean) of 14.4 years of education compared to a mean of 13.6 years of education for the dropout mothers. This difference was statistically reliable, with a probability of less than .05 (5 percent, or one in 20) that the difference found was just due to random chance or sampling error.

The M means "mean," or arithmetic mean. Add up all the scores and divide by the number of scores for the mean number. It is a very common and handy summary statistic.

They also had higher income, with the first group making 3.6 times the poverty-level wage (on the average) compared to 3.2 times poverty wages for the dropout group; and were more likely to have a man at home, 85% vs. 76%.

The use of a p-level, such as p < .05 or p < .01, is very common in social science research. You can have confidence that the result is reliable, or is a genuine difference between groups, because it is unlikely to have arisen by chance. The .05 or .01 or .001 tells you that the result could have arisen by chance, once in 20 experiments, or once in 100, or once in 1,000. So you can conclude that probably it is a genuine effect.

The p-value or p-level is sometimes called statistical significance, but that is a little misleading, since it sounds as if it means importance. It doesn't. It just means that the result is more or less unlikely to have arisen if there was in fact no difference between the group. The difference itself could be trivial. Is 3.6 times the poverty level really much richer than 3.2 times? Is it an important difference between those groups? That isn't really answered by the p-value alone.

Generally, any result whose p-value is not p < .05 or better is not considered statistically significant. But statistically non-significant results could still be very important differences. Maybe 85% vs. 76% isn't significantly different in this study, but for the women and children without a man in the house, it could be very important, and may explain why there is a difference in incomes.
_________________
You live a new life for every new language you speak. -Czech proverb

[tisogai]

Hello, CP

THANK YOU SO MUCH! I really appreciate your time spent on your two replies. I read every word of them and can now understand that very well. Your answers made all very clear.

I'm in the education field. This hard thing is just for my personal interest...

I hope you don't mind helping me out a bit more on this same topic. I'm posting another question, so would you please take a look at that?

Thanks again.