SLP 510 Homework #5 & 6 Questions

Nominal

In this level of measurement, researchers assign participants and their responses to categories such as male or female, type of utterance, type of audiogram, type of aphasia or other disorder, dialect of American English, and so forth.

Ordinal

In this level of measurement, researchers rank participants and observations from high to low relative to one another or researchers use a rating scale.

Interval

This level of measurement provides information about which participants have higher or lower scores, as well as by how much participants differ.

Ratio

This level of measurement provides information about which participants have higher and lowers scores, AND can be arranged on a continuum, differences, and there is an absolute zero.

independent samples t test

Researchers are studying the relationship between two treatment approaches. Each treatment had a group of 20 clients receive treatment. To determine if the group outcomes were different, which statistical test would be most appropriate for analyzing the data?

correlational analysis

Researchers are studying the relationship between two variables: the amount of therapy homework completed (focusing on sentence structure) with correct sentence structure in the classroom setting. They could perform what to determine strength of the relationship between those variables?

paired t test

A research study compares the mean number of dysfluencies in a reading passage before and after a six-week fluency program for the same group of 10 teenagers. Which statistical test would be most appropriate for analyzing the data?

Repeated measure ANOVA

A research study measures scores on a functional communication scale for a 10 patients in a medical facility. Each patient is scored with how they communicate with three different types of communication partners (spouse/close family member, known medical staff, unknown medical staff). What is the best statistical test for comparing the results of patients across different communication partners?

type II error

A researcher conducts a single-subject study to determine whether a new phonological intervention improves speech intelligibility compared to the standard treatment. The results show no statistically significant difference, so the researcher concludes that the new treatment is not more effective. Which type of error might the researcher be making in this situation?

standard deviation

This reflects the dispersion of scores around the mean and most commonly used measure of variability

ratio data

A level of data measurement that has three characteristics: (a) the ability to arrange numbers on a continuum, (b) the ability to specify amount and differences between amounts, and (c) the ability to identify an absolute zero relative to a characteristic.

inference

When researchers study a sample from a large population, they still want to draw conclusions about the population as a whole

maturation threat

Increases in performance that are due to the participants' growth and development over time

parameters

In group studies, findings that are numerical summaries that come from observations on the entire population are called

volunteerism

A source of bias that cannot be avoided because the process of informed consent specifies that all persons who participate in research should do so on a voluntary basis. However, persons who volunteer to participate in research might differ from the general population in some systematic way.

mode

Based on frequency information and is the category, response, or score that occurs most often. primary measure of central tendency for nominal measures

margin of error

Reflects the potential distribution of means, the sampling distribution, that would emerge if you took many samples from a population and calculated a mean for each of those samples

sample of convenience

This refers to using a group of participants who are easy to access

mortality threat

Some participants drop out before the end of a study

type II error

The type of error that fails to reject the null hypothesis when it is, in fact, an incorrect hypothesis