Parametric Statistics

# Putting it all together

In research, we often collect information from a collection of individuals that are drawn from a larger group. Since the procedures for information collection have real costs, researchers are forced to make the assumption that the individuals selected for study are a true representation of all of the individuals within the larger group. In quantitative analyses the larger group is the population, (represented by the letter ‘N’), and the selected subgroup is referred to as the sample (represented by the letter ‘n’). Given the assumption that the sample represents the population from which it was drawn, then it is also assumed that any computations, estimates, or inferences based on the measures from the sample, must also represent the population from which the sample was selected.

As such, the average score computed for the sample is assumed to represent the average score for the population. Likewise, the variability of scores within the sample (the subgroup) is expected to represent the variability of the scores within the population (the larger group). Similarly, the standardized estimate of the differences computed for the sample should represent the standardized estimate of differences computed for the population.

The confidence interval is based on the following relationship between the sample means and the true population mean or [µ]:

the lower limit of the sample mean < µ < upper limit of the sample mean

This sentence is read as: The lower limit of the sample mean is less than the true population estimate which is less than the upper limit of the sample mean. Therefore: the population mean = sample mean ± sampling error : µ =  $\overline{x}$ ± (SE),

where:

• µ refers to the measure of central tendency for the population
• is the sample mean and refers to the measure of centrality in a sample
• Standard Error (SE) — error due to randomness

There are two basic assumptions in this approach:

First, we assume that the sample mean is only our best estimate of the true population mean.

Second, we assume that the chance associated with the sample mean’s ability to represent the true population mean is dependent upon the ability of the sample scores to represent the population scores.

So that by adding or subtracting the sampling error to or from the sample mean we will be able to identify the range within which the true population estimate falls.

## Standard Error

The term sampling error, which is also called the standard error of the mean, is a measure of the extent to which the sample means can be expected to vary due to chance. In other words, the standard error of the mean provides an estimation of the variance (or error) of the mean in the sample and can be attributed to the sampling characteristics associated with the sample.

## The Confidence Interval (95%)

Confidence intervals help the researcher determine the accuracy that a sample estimate represents a true population parameter. In most studies, the researcher has an implicit expectation that the sample is representative of the larger group (i.e. the population). Therefore, if we assume that our sample represents a population, then we must also assume that any computations, estimates, or inferences based on the numbers from the sample, must also represent the population from which the sample was selected. As such, the average score computed for the sample is assumed to represent the average score for the population; similarly, the variability of scores within the sample (the subgroup) should represent the variability of the scores within the population (the larger group).

Given that the sample is an accurate representation of the population, standardized estimates of the differences computed for the sample should represent the standardized estimates of differences computed for the population. One can also expect that the measure of central tendency for the population (µ) can only be estimated by the measure of central tendency for the sample, whereby = µ and therefore it is accepted that there will always be some amount of error due to known and unknown factors. While the sample mean, variance and standard deviation each represent estimates of the true population values, the value that represents the accuracy of our projected estimates are expressed as a measure of confidence. We can, therefore, assume that the confidence interval is an accurate representation of the actual space within which we could expect to find the true population measures. Such expectations are based on the following principles: we assume that the sample mean is only our best estimate of the true population mean. We assume that the chance associated with the sample mean’s ability to represent the true population mean is dependent upon the ability of the sample scores to represent the population scores. By adding or subtracting the sampling error to or from the sample mean we will be able to identify the range within which the true population estimate falls.

The term sampling error refers to the errors that occur in the process of data collection. Sampling error is expected and should thus be accounted for in the computation of the estimates that represent the data. Researchers state that the estimates (measures of central tendency, frequencies, or ratio estimates) produced from a selected sample are expected to represent the true population estimate within a specific range. For example, the researcher states that: They are 95% confident that the sample mean represents the true population mean within 10% error.

Typically, researchers indicate that they would like to be at least 95% confident that the sample mean is an estimate of the population mean. Therefore, the researcher is suggesting that 19 out of 20 times the sample mean ± sampling error will include [µ]. The confidence interval is based on the following relationship between the sample mean and the true population mean or [µ]: This sentence is read as: The lower limit of the sample mean is less than the true population estimate which is less than the upper limit of the sample mean.

The standard error of the mean, is computed by the following formula shown here: $s.e. = {s \over{\sqrt{n}}}$

Example: Compute the confidence interval at 95% for a given sample mean where the mean = 58 ± standard deviation (s) = 13 and n=25 participants. To compute the standard error (se) we use: $s.e. = {s \over{\sqrt{n}}}$ = $s.e. = {13 \over{\sqrt{25}}}$ = 2.6

The upper and lower limit of the 95% confidence interval for the mean = 58 is computed with the basic formula: $CI_{95} = { \overline{x} \pm }(1.96 \times{se})$

$58 \pm [1.96 \times{2.6}] {\rightarrow}\textit{95% confidence interval is 58} \pm 5.1$. Which means that there is a 95% probability or chance that the range 52.9 and 63.1 will capture the true population mean µ.

## THE BASIC PREMISE OF ESTIMATION AND CONFIDENCE INTERVALS

µ = sample mean ± (1.96 x standard error of the mean)

where:

• µ refers to the measure of central tendency for the population
• sample mean refers to the measure of central tendency for the sample
• standard error of the mean also known as sampling error is the estimate by which the mean can vary

In computing confidence intervals, we determine the estimate of the error of the sample selected or the sampling error. This error is also called the standard error of the mean and is a measure of “the extent to which the sample means can be expected to vary due to chance”. In other words, the standard error of the mean is “an estimate of the error associated with the observed mean in this specific sample” and is due to the sampling characteristics associated with this sample.