Topic: Does Our Personality Remain Stable Over Time?
Statistical concepts covered:
In this applet, you’ll expand on your knowledge of research design, specifically longitudinal studies, and the concept of standard error (defined in the first Statistical Lesson) and the impact of the size of the sample.
Longitudinal studies track individuals over a period of time, usually many years, to determine change. These types of studies provide great insight in regards to how we change as we grow or get older. Longitudinal research was conducted by the National Survey of Midlife Development in the United States in 1995-1996 and again in 2004-2005 for a variety of health and well-being measures. One of the areas measured was personality, including what we refer to as the Big Five dimensions of personality: Openness to experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. In this applet we will utilize the data collected from this study to assess the impact of sample size and time on personality and standard error.
Statistical Lesson: Our goal with most statistics is to make guesses about the population using what we learn from a sample. When we calculate the sample mean or average, we can also determine how close this average might be to the actual population average. We estimate this via the standard error, which is how much our sample average deviates from the population average. If the standard error is small, then we are more confident that the sample average is closer to the population average. However, if the standard error is large, then we are less confident that our sample can accurately predict the population average. Since our goal is typically to make assumptions about the population using the sample, then it is best to reduce the size of our standard error.
The sample size is important because the larger the sample, the better it reflects the population. Smaller data sets can have high levels of variability and may be misleading. Fortunately, we don’t necessarily need to sample everyone or everything in the population. Moderately sized data sets are typically sufficient and more realistic to achieve.
Having larger sample sizes decreases the variability and standard error, allowing us to better predict the population values. However, which of the following is a reason why we do NOT include the entire population in a sample? Select all that apply.
|bI0LPa9lfHQ+dYqk||It is not realistic to sample everyone in a population of interest, especially if that population is extremely large or difficult to reach.|
|bI0LPa9lfHQ+dYqk||It is not necessary to sample everyone in a population of interest as long as our sample is representative of the population.|
|bI0LPa9lfHQ+dYqk||It might be difficult to determine every individual in the population who should be included in the sample.|
Congratulations! You have completed this activity.