Wealth risks while working

Briefly describe a time series that would allow you to investigate the impact of your
intervention, and explain how you would collect and record the additional data. Be
B) Select a method for geographically representing the provided data. Use the method
to create a graph.
C) Analyze the data and provide a narrative description of your analysis.

Conducting Time Series Research

The purpose of this intervention is to counter the effects of exposure to health risks
while working in the factory. Quasi experimental design is used to determine the
effectiveness of introduction of gas masks for welders. I will use the most basic form of this
design which involves both pre-intervention and post-intervention part. This method is
simple as it relies on a null hypothesis which assumes a small change after the intervention.
Data collection for this time series research will solely come from the medical
records. The use of data archives is one of the approaches used in time series research. Since
the intervention is conducted by another party and not solely for the experiment, the
researcher is free to use such information for analysis (Glass). Welders usually visit the
dispensary several times in a month. At the dispensary, the clinician does a number of tests to
check the levels of gas fumes in the blood. I will collect the relevant data from the dispensary
on day 1 and subsequent days.
In recording the data for this research, the point in time aspect of a time series design
comes into play. This is the basic unit of analysis in this series and refers to day, hour, month,
and so on (McCleary & McDowall 2012). A time series can be either discrete or continuous,

Time Series Research 2

and in this case, I choose to use the continuous approach. In this method, the data from the
dispensary will be collected daily from day 1 to day 60. Therefore, there will be no time
between these periods, which is referred to as lag in the discrete method.

How to Represent the Data

Concentration in %


20 40 60


The data in this case is an example of an interrupted time series test. For 60
consecutive days 1 data of the welder is collected from the dispensary database. On the first
day, I will collect data for all the welders, on the second day the same, and so on until the 60
days are over.

Time Series Research 3

In this experiment, I am using the change in fumes concentration in the blood and
lungs as an indication of the success or failure of the intervention. If the fumes increase after
the intervention, then the intervention is considered as unsuccessful. Similarly, a decrease in
concentration of fumes components in the results indicates success of the intervention.
Analysis of the Data
The figure above shows the rate of fumes concentration for the entire 60 days of this
experiment. From day 1 to day 20, the data was collected from a welder who did not use gas
mask during his welding. After 20 days of data collection and recording, gas masks were
already in use and I was able to take data for welder who used the equipment. At this point, I
assumed that the welder (human subject) was under behavioral modification aimed at
suppressing fumes content in blood and lungs. To cater for the termination segment of a
quasi-experimental analysis, the welding mask was withdrawn for a further 20 days. During
the entire period, I continued collecting data from the dispensary( Kratochwill & Levin,
By collecting the data in three segments, I wanted to ascertain the effect of
introduction of the gas masks for welders. One might want to look at the specific times when
the no-treatment shifts to treatment. One may also learn something about the introduction of
gas masks by looking the point where the gas mask is withdrawn. In my experiment, the
introduction of gas masks is the intervention while fumes concentration represents the
outcome (McCleary & McDowall 2012).

Time Series Research 4

A time series experiment offers some very interesting insights depending on the
behavior of the dependent variable in relation to the intervention point. If there is a sudden
shift of the variable at that point, then you can make a conclusion that the intervention did
affect the dependent variable. In this experiment, a major question would be: “Did the
introduction of gas masks reduce the amount of gas fumes inhaled by the welders?”
After the introduction of gas masks after 20 days, the fumes concentration reduced.
After the withdrawal of the same on day 41, the concentration started to rise again.
It is very important to have an idea about how the results of this experiment might be,
considering extraneous influences are likelihood. In detailed experiments of this nature, the
graph shows ephemeral jumps which might lead an inexperienced researcher astray. It is
worthwhile to note that graphs changes in a time series research may be due to varied factors,
others far from the intervention.
In this and any other time series experiment, it is necessary to determine whether you
are dealing with a stationary or a non-stationary time series (Glass). Since the graph of this
experiment changes distinctively around day 20 and day 40, the conclusion is that it is
stationary time series. That is why it was so easy to detect the effects of the introduction of
gas mask, which is the intervention in the experiment.

Time Series Research 5


Glass, G. V. (n.d.). Interrupted time series quasi-experiments. (Master’s thesis, Arizona State
Kratochwill, T. R. & Levin, J.R. (Eds.). (1992). Single-case research design and
analysis: New directions for psychology and education. N.Y.: Lawrence